├── terms ├── The Three Laws of Robotic.md ├── Pseudo AI.md ├── Agentic Reasoning.md ├── Misaligned AI.md ├── Pasted image 20240416220310.png ├── Labelled Data.md ├── Gptq.md ├── Adapters.md ├── Awq.md ├── Completions.md ├── Exui.md ├── Georgi Gerganov.md ├── Chatui.md ├── Guanaco (Model).md ├── Koboldcpp.md ├── Mikupad.md ├── Mistral 7b Instruct.md ├── Phixtral.md ├── Tagging.md ├── Unsloth.md ├── Code Llama.md ├── Code Models.md ├── Mergekit.md ├── Prompt 1.md ├── Training Data.md ├── Whisper.md ├── ChatGPT.md ├── Lemma.md ├── Metadata.md ├── Tri Dao.md ├── Mistral 7b.md ├── Tinyllama.md ├── Trl.md ├── Actionable Intelligence.md ├── Lm Studio.md ├── Peft.md ├── Tim Dettmers.md ├── Flash Attention.md ├── Gpt-4.md ├── Ipo.md ├── Windowing.md ├── Big Code Models Leaderboard.md ├── Domain Knowledge.md ├── Generalized Model.md ├── Goliath-120b.md ├── Rope.md ├── Bigcode.md ├── Flash Attention 2.md ├── Gguf.md ├── Llava.md ├── Local Llms.md ├── MLX.md ├── Notus.md ├── Prompt Chaining.md ├── Truthfulqa.md ├── Wizardcoder.md ├── Cognitive Computations.md ├── Frankenmerge.md ├── Nous Research.md ├── Open Llm Leaderboard.md ├── Ppo.md ├── The Stack.md ├── Zero Shot Extraction.md ├── Thebloke.md ├── Transfer Learning.md ├── Averaging.md ├── Category.md ├── Metacontext And Metaprompt.md ├── Text Gen. Ui, Inference Engines.md ├── Customdomain Language Model.md ├── Dpo.md ├── Extraction Or Keyphrase Extraction.md ├── Chatbot Arena.md ├── Exl2.md ├── Gpt.md ├── Llm.md ├── Localllama.md ├── Mixtral.md ├── Oobabooga Text-Generation-Webui.md ├── Parsing.md ├── Symbolic Ai.md ├── Transformer.md ├── Axolotl.md ├── Fine-Tuned Model.md ├── Generative Ai (Genai).md ├── Ggml.md ├── Humaneval.md ├── Hyperparameters.md ├── Data Drift.md ├── Etl (Entity Recognition, Extraction).md ├── Model Merging.md ├── Data Scarcity.md ├── Multimodal.md ├── Reinforcement Learning With Human Feedback (Rlhf).md ├── Superhot.md ├── Annotation.md ├── Number Of Parameters In Models.md ├── Roai.md ├── Vicuna.md ├── Auto-Complete.md ├── Content.md ├── Convolutional Neural Networks (Cnn).md ├── Data Discovery.md ├── Hallucitations.md ├── Supervised Learning.md ├── Lexicon.md ├── Few-Shot.md ├── transformers.md ├── Instruct-Tuning.md ├── Lima.md ├── Tokens.md ├── Laser.md ├── Nlq (Aka Natural Language Query).md ├── Recurrent Neural Networks (Rnn).md ├── Sentiment.md ├── Zephyr.md ├── Bagel.md ├── Data Ingestion.md ├── Extractive Summarization.md ├── Kto.md ├── Multitask Prompt Tuning (Mpt).md ├── Benchmark.md ├── Semantic Network.md ├── Tunable.md ├── Mt-Bench.md ├── Qlora.md ├── Simple Knowledge Organization System (Skos).md ├── Syntax.md ├── Artificial Neural Network (Ann).md ├── Hallucination.md ├── Prompt Engineering.md ├── Reward Model.md ├── Data Extraction.md ├── Knowledged Based Ai.md ├── Phi 2.md ├── Random Forest.md ├── Triple or Triplet Relations aka (Subject Action Object (SAO).md ├── Alpaca.md ├── Did You Mean (Dym).md ├── Few-Shot Learning.md ├── Dpo Overfits.md ├── Edge Model.md ├── Overfitting.md ├── Plugins.md ├── Self-Supervised Learning.md ├── Speech Recognition.md ├── Anaphora.md ├── ControlNet.md ├── Knowledge Graphs.md ├── Langops (Language Operations).md ├── Language Data.md ├── Temperature.md ├── Auto-Classification.md ├── Context Length.md ├── Data Labelling.md ├── Multimodal Models And Modalities.md ├── Training Set.md ├── Sillytavern (Aka St).md ├── Cataphora.md ├── Post-Processing.md ├── Prompt.md ├── Categorization.md ├── Content Enrichment Or Enrichment.md ├── Knowledge Engineering.md ├── Thesauri.md ├── Moe Merging.md ├── Morphological Analysis.md ├── Nlt (Aka Natural Language Technology).md ├── Pemt (Aka Post Edit Machine Translation).md ├── Rules-Based Machine Translation (Rbmt).md ├── Entity.md ├── Category Trees.md ├── Conversational Models.md ├── Environmental, Social, And Governance (Esg).md ├── Rl.md ├── Specialized Corpora.md ├── Uncensored Models.md ├── Accuracy.md ├── Llama 2.md ├── Emotion Ai (Aka Affective Computing).md ├── Explainable Aiexplainability.md ├── Hugging Face (Hf).md ├── Test Set.md ├── Composite Ai.md ├── Lora.md ├── Unstructured Data.md ├── Knowledge Model.md ├── Pretraining.md ├── Classification.md ├── Cognitive Map.md ├── Pretrained Model.md ├── Linked Data.md ├── Model.md ├── Speech Analytics.md ├── Computational Linguistics (Text Analytics, Text Mining).md ├── Corpus.md ├── Structured Data.md ├── Grounding.md ├── Nlg (Aka Natural Language Generation).md ├── Semantic Search.md ├── Botshit.md ├── Disambiguation.md ├── Foundational Model.md ├── Peft (Parameter-Efficient Fine-Tuning).md ├── Subject-Action-Object (SAO).md ├── Hybrid Ai.md ├── Mistral.md ├── Treemap.md ├── Computational Semantics (Semantic Technology).md ├── Model Parameter.md ├── Bert (Aka Bidirectional Encoder Representation From Transformers).md ├── Large Language Models (Llm).md ├── Evals.md ├── Semantics.md ├── Intelligent Document Processing (Idp) Or Intelligent Document Extraction And Processing (Idep).md ├── Relations.md ├── Tuning (Aka Model Tuning Or Fine Tuning).md ├── Logits.md ├── Deep Learning.md ├── Natural Language Understanding.md ├── Sentiment Analysis.md ├── Conversational Ai.md ├── Insight Engines.md ├── Model Drift.md ├── Perplexity.md ├── Similarity (And Correlation).md ├── Text Summarization.md ├── Generative Summarization.md ├── Summarization (Text).md ├── Text Analytics.md ├── Pre-Processing.md ├── WizardLM 2.md ├── Alice and Bob.md ├── Imatrix.md ├── Zero-Shot.md ├── Base Vs Conversational.md ├── Ai Feedback (Aif).md ├── Auto-Regressive.md ├── Natural Language Processing.md ├── Question & Answer (Q&A).md ├── Co-Occurrence.md ├── Ontology.md ├── Semi-Structured Data.md ├── Pre-Training.md ├── Quantization.md ├── Retrieval Augmented Generation (Rag).md ├── Gqa.md ├── Symbolic Methodology.md ├── Total Parameters.md ├── Tokenizer.md ├── Part-Of-Speech Tagging.md ├── Responsible Ai.md ├── Rlhf (Reinforcement Learning With Human Feedback).md ├── Controlled Vocabulary.md ├── Definition.md ├── Finetunes Fine-Tuning.md ├── Cosine Similarity.md ├── Recall.md ├── Llama.md ├── Llama 3.md ├── Inference Inferencing.md ├── Precision.md ├── Machine Learning (Ml).md ├── Taxonomy.md ├── F-Score (F-Measure, F1 Measure).md ├── Knowledge Graph.md ├── Agentic Al Design Patterns.md ├── Model Reliability.md ├── Effective Accelerationism.md ├── Effective Altruism.md ├── Red-Teaming.md ├── Alignment Training.md ├── Indexer (LLM).md ├── AI-Washing.md ├── Trustworthy AI.md ├── Agentic Workflows.md ├── RLHF (Reinforcement Learning From Human Feedback).md ├── AI Alignment.md └── AGI (Artificial General Intelligence).md ├── cover art └── azai-2024-cover-art-600x400px.png ├── CONTRIBUTING.md ├── README.md └── LICENSE /terms/The Three Laws of Robotic.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | ### Footnotes -------------------------------------------------------------------------------- /terms/Pseudo AI.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | See [AI-Washing](./AI-Washing.md). -------------------------------------------------------------------------------- /terms/Agentic Reasoning.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | See [Agentic Workflows](./Agentic%20Workflows.md) -------------------------------------------------------------------------------- /terms/Misaligned AI.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | See [AI Alignment](./AI%20Alignment.md) 5 | 6 | ### Footnotes -------------------------------------------------------------------------------- /terms/Pasted image 20240416220310.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/alexfazio/ai-glossary/main/terms/Pasted image 20240416220310.png -------------------------------------------------------------------------------- /cover art/azai-2024-cover-art-600x400px.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/alexfazio/ai-glossary/main/cover art/azai-2024-cover-art-600x400px.png -------------------------------------------------------------------------------- /terms/Labelled Data.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | see Data Labelling. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Gptq.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A popular quantization technique. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Adapters.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Another popular library to do PEFT. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Awq.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Another popular quantization technique. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Completions.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The output from a generative prompt. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Exui.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | An open-source UI to use open-source models. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Georgi Gerganov.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The creator of llama.cpp and ggml! 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Chatui.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | An open-source UI to use open-source models. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Guanaco (Model).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A LLaMA fine-tune using QLoRA tuning. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Koboldcpp.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | An open-source UI to use open-source models. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Mikupad.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | An open-source UI to use open-source models. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Mistral 7b Instruct.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A fine-tuned version of Mistral 7B. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Phixtral.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A MoE merge of Phi 2 DPO and Dolphin 2 Phi 2. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Tagging.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | See Parts-of-Speech Tagging (aka POS Tagging). 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Unsloth.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A higher-level library to do PEFT (using QLoRA) 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Code Llama.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The best base code model. It's based on Llama 2. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Code Models.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | LLMs that are specifically pre-trained for code. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Mergekit.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A cool open-source tool to quickly merge repos. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Prompt 1.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A phrase or individual keywords used as input for GenAI. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Training Data.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The collection of data used to train an AI model. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Whisper.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The state-of-the-art speech-to-text open-source model. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/ChatGPT.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | RLHF-finetuned GPT-3 model that is very good at conversations. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Lemma.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The base form of a word representing all its inflected forms. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Metadata.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Data that describes or provides information about other data. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Tri Dao.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The author of both techniques and a legend in the ecosystem. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Mistral 7b.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A pre-trained model trained by Mistral. Released via torrent. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Tinyllama.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A project to pre-train a 1.1B Llama model on 3 trillion tokens. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Trl.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A library that allows to train models with DPO, IPO, KTO, and more! 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Actionable Intelligence.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Information you can leverage to support decision making. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Lm Studio.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A nice advanced app that runs models on your laptop, entirely offline. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Peft.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A popular OS library to do PEFT! It's used in other projects such as trl. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Tim Dettmers.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A researcher that has done a lot of work on PEFT and created QLoRA. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Flash Attention.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | An approximate attention algorithm which provides a huge speedup. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Gpt-4.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A kinda good model, but we don't know what it is. The rumors say it's a MoE. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Ipo.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A change in the DPO objective which is simpler and less prone to overfitting. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Windowing.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A method that uses a portion of a document as metacontext or metacontent 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Big Code Models Leaderboard.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A leaderboard to compare code models in the HumanEval dataset. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Domain Knowledge.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The experience and expertise your organization has acquired over time. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Generalized Model.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A model that does not specifically focus on use cases or information. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Goliath-120b.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A frankenmerge that combines two Llama 70B models to achieve a 120B model 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Rope.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A technique that allows you to significantly expand the context lengths of a model. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Bigcode.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | An open scientific collaboration working in code-related models and datasets. BitNet: 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Flash Attention 2.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | An upgrade to the flash attention algorithm that provides even more speedup. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Gguf.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A format introduced by llama.cpp to store models. It replaces the old file format, GGML. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Llava.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A multimodal model that can receive images and text as input and generate text responses. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Local Llms.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | If we have models small enough, we can run them in our computers or even our phones! 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/MLX.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A new framework for Apple devices that allows easy inference and fine-tuning of models. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Notus.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A trained variation of Zephyr but with better-filtered and fixed data. It does better! 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Prompt Chaining.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | An approach that uses multiple prompts to refine a request made by a model. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Truthfulqa.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A not-so-great benchmark to measure a model's ability to generate truthful answers. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Wizardcoder.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A code model released by WizardLM. Its architecture is based on Llama. WizardLM 2: 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Cognitive Computations.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A community (led by Eric Hartford) that is fine-tuning a bunch of models. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Frankenmerge.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | It allows to concatenate layers from different LLMs, allowing you to do crazy things. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Nous Research.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | An open-source Discord community turned company that releases a bunch of cool models. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Open Llm Leaderboard.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A leaderboard where you can find benchmark results for many open-access LLMs. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Ppo.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A type of reinforcement learning algorithm that is used to train a model. It is used in RLHF. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/The Stack.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A dataset of 6.4TB of permissible-licensed code data covering 358 programming languages. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Zero Shot Extraction.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The ability to extract data from text with no previous training or annotations. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Thebloke.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A bloke that quantizes models. As soon as a model is out, he quantizes it! See their HF Profile. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Transfer Learning.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A technique in which a pretrained model is used as a starting point for a new ML task. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Averaging.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The most basic merging technique. Pick two models, average their weights. Somehow, it kinda works! 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Category.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A category is a label assigned to a document in order to describe the content within said document. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Metacontext And Metaprompt.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Foundational instructions on how to train the way in which the model should behave. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Text Gen. Ui, Inference Engines.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | If you don't know how to code, there are a couple of tools that can be useful. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Customdomain Language Model.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A model built specifically for an organization or an industry – for example Insurance. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Dpo.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A type of training which removes the need for a reward model. It simplifies significantly the RLHF-pipeline. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Extraction Or Keyphrase Extraction.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Mutiple words that describe the main ideas and essence of text in documents. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Chatbot Arena.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A popular crowd-sourced open benchmark of human preferences. It's good to compare conversational models. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Exl2.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A different quantization format used by a library called exllamav2 (among many others) which has variable bitrates. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Gpt.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A type of transformer that is trained to predict the next token in a sentence. GPT-3 is an example of a GPT model. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Llm.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A Large Language Model. Usually a transformer-based model with a lot of parameters...billions or even trillions. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Localllama.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A Reddit community of practitioners, researchers, and hackers doing all kinds of crazy things with ML models. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Mixtral.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A MoE model developed by Mistral AI, incorporating the MoE architecture to optimize efficiency and performance. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Oobabooga Text-Generation-Webui.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A simple web app that allows you to use models without coding. It's very easy to use! 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Parsing.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Identifying the single elements that constitute a text, then assigning them their logical and grammatical value. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Symbolic Ai.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Add to Symboilic Methodology parthethetically so it looks like this: “Symbolic Methodology (Symbolic AI)” 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Transformer.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A type of neural network architecture that is very good at language tasks. It is the basis for most LLMs. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Axolotl.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A cute animal that is also a high-level tool to streamline fine-tuning, including support for things such as QLoRA. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Fine-Tuned Model.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A model focused on a specific context or category of information, such as a topic, industry or problem set 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Generative Ai (Genai).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | AI techniques that learn from representations of data and model artifacts to generate new artifacts. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Ggml.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A tensor library in ML, allowing projects such as llama.cpp and whisper.cpp (not the same as GGML, the file format). 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Humaneval.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A very small dataset of 164 Python programming problems. It is translated to 18 programming languages in MultiPL-E. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Hyperparameters.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | These are adjustable model parameters that are tuned in order to obtain optimal performance of the model. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Data Drift.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Data Drift occurs when the distribution of the input data changes over time; this is also known as covariate shift. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Etl (Entity Recognition, Extraction).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Entity extraction is an NLP function that serves to identify relevant entities in a document. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Model Merging.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A technique that allows us to combine multiple models of the same architecture into a single model. Read more here. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Data Scarcity.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The lack of data that could possibly satisfy the need of the system to increase the accuracy of predictive analytics. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Multimodal.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A single model that can handle multiple modalities. For example, a model that can generate text and images at the same time. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Reinforcement Learning With Human Feedback (Rlhf).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A ML algorithm that learns how to perform a task by receiving feedback from a human. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Superhot.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A technique that allows expanding the context length of RoPE-based models even more by doing some minimal additional training. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Annotation.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The process of tagging language data by identifying and flagging grammatical, semantic or phonetic elements in language data. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Number Of Parameters In Models.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Refers to the total count of adjustable elements within a model, crucial for learning and error minimization. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Roai.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Return on Artificial Intelligence (AI) is an abbreviation for return on investment (ROI) on an AI-specific initiative or investment. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Vicuna.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A cute animal that is also a fine-tuned model. It begins from LLaMA-13B and is fine-tuned on user conversations with ChatGPT. VRAM: 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Auto-Complete.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Auto-complete is a search functionality used to suggest possible queries based on the text being used to compile a search query.. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Content.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Individual containers of information — that is, documents — that can be combined to form training data or generated by Generative AI. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Convolutional Neural Networks (Cnn).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A deep learning class of neural networks with one or more layers used for image recognition and processing. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Data Discovery.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The process of uncovering data insights and getting those insights to the users who need them, when they need them. Learn more 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Hallucitations.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Made up data that include fabricated, inaccurate or misaligned references or sources that are presented as fact in generated text. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Supervised Learning.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | An ML algorithm in which the computer is trained using labeled data or ML models trained through examples to guide learning. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Lexicon.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Knowledge of all of the possible meanings of words, in their proper context; is fundamental for processing text content with high precision. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Few-Shot.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A type of prompt that is used to generate text with fine-tuning. We provide a couple of examples to the model. This can improve the quality a lot! 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/transformers.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A Python library to access models shared by the community. It allows you to download pre-trained models and fine-tune them for your own needs. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Instruct-Tuning.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A type of fine-tuning that uses instructions to generate text ending in more controlled behavior in generating responses or performing tasks. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Lima.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A model that demonstrates strong performance with very few examples. It demonstrates that adding more data does not always correlate with better quality. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Tokens.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A unit of content corresponding to a subset of a word. Tokens are processed internally by LLMs and can also be used as metrics for usage and billing. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Laser.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A technique that reduces the size of the model and increases its performance by reducing the rank of specific matrices. It requires no additional training. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Nlq (Aka Natural Language Query).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A natural language input that only includes terms and phrases as they occur in spoken language (i.e. without non-language characters). 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Recurrent Neural Networks (Rnn).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A neural network model commonly used in natural language process and speech recognition allowing previous outputs to be used as inputs. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Sentiment.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Sentiment is the general disposition expressed in a text. Read our blog post “Natural Language Processing and Sentiment Analysis” to learn more about sentiment. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Zephyr.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A 7B Mistral-based model trained with DPO. It has similar capabilities to the Llama 2 Chat model of 70B parameters. It came out with a nice handbook of recipes. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Bagel.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A process which mixes a bunch of supervised fine-tuning and preference data. It uses different prompt formats, making the model more versatile to all kinds of prompts. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Data Ingestion.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The process of obtaining disparate data from multiple sources, restucturing it, and importing it into a common format or repository to make it easy to utilize. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Extractive Summarization.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Identifies the important information in a text and groups the text fragments together to form a concise summary. Also see Generative Summarization 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Kto.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | While PPO, DPO, and IPO require pairs of accepted vs rejected generations, KTO just needs a binary label (accepted or rejected), hence allowing to scale to much more data. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Multitask Prompt Tuning (Mpt).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | An approach that configures a prompt representing a variable — that can be changed — to allow repetitive prompts where only the variable changes. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Benchmark.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A benchmark is a test that you run to compare different models. For example, you can run a benchmark to compare the performance of different models on a specific task. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Semantic Network.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A form of knowledge representation, used in several natural language processing applications, where concepts are connected to each other by semantic relationship. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Tunable.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | An AI model that can be easily configured for specific requirements. For example, by industry such as healthcare, oil and gas, departmental accounting or human resources. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Mt-Bench.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A multi-turn benchmark of 160 questions across eight domains. Each response is evaluated by GPT-4. (This presents limitations...what happens if the model is better than GPT-4?) 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Qlora.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A technique that combines LoRAs with quantization, hence we use 4-bit quantization and only update the LoRA parameters! This allows fine-tuning models with very GPU-poor GPUs. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Simple Knowledge Organization System (Skos).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A common data model for knowledge organization systems such as thesauri, classification schemes, subject heading systems, and taxonomies. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Syntax.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The arrangement of words and phrases in a specific order to create meaning in language. If you change the position of one word, it is possible to change the context and meaning. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Artificial Neural Network (Ann).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Commonly referred to as a neural network, this system consists of a collection of nodes/units that loosely mimics the processing abilities of the human brain. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Hallucination.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | When a model generates responses that may be coherent but are not actually accurate, leading to the creation of misinformation or imaginary scenarios...such as the one above! 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Prompt Engineering.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The craft of designing and optimizing user requests to an LLM or LLM-based chatbot to get the most effective result, often achieved through significant experimentation. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Reward Model.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A model that is used to generate rewards. For example, you can train a model to generate rewards for a game. The model will learn to generate rewards that are good for the game! 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Data Extraction.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Data extraction is the process of collecting or retrieving disparate types of data from a variety of sources, many of which may be poorly organized or completely unstructured. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Knowledged Based Ai.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Knowledge-based systems (KBs) are a form of artificial intelligence (AI) designed to capture the knowledge of human experts to support decision-making and problem-solving. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Phi 2.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A pre-trained model by Microsoft. It only has 2.7B parameters but it's quite good for its size! It was trained with very little data (textbooks) which shows the power of high-quality data. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Random Forest.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A supervised machine learning algorithm that grows and combines multiple decision trees to create a “forest.” Used for both classification and regression problems in R and Python. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Triple or Triplet Relations aka (Subject Action Object (SAO).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | An advanced extraction technique which identifies three items (subject, predicate and object) that can be used to store information. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Alpaca.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A dataset of 52,000 instructions generatd with OpenAI APIs. It kicked off a big wave of people using OpenAI to generate synthetic data for instruct-tuning. It costed about $500 to generate. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Did You Mean (Dym).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | “Did You Mean” is an NLP function used in search applications to identify typos in a query or suggest similar queries that could produce results in the search database being used. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Few-Shot Learning.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | In contrast to traditional models, which require many training examples, few-shot learning uses only a small number of training examples to generalize and produce worthwhile output. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Dpo Overfits.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Although DPO shows overfitting behaviors after one behavior, it does not harm downstream performance on chat evaluations. Did your ML teachers lie to us when they said overfitting was bad? 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Edge Model.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A model that includes data typically outside centralized cloud data centers and closer to local devices or individuals — for example, wearables and Internet of Things (IoT) sensors or actuators. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Overfitting.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Occurs in ML when a model learns the training data too well, capturing noise and specific patterns that do not generalize to new, unseen data, leading to poor performance on real-world tasks. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Plugins.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A software component or module that extends the functionality of an LLM system into a wide range of areas, including travel reservations, e-commerce, web browsing and mathematical calculations. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Self-Supervised Learning.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | An approach to ML in which labeled data is created from the data itself. It does not rely on historical outcome data or external human supervisors that provide labels or feedback. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Speech Recognition.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Speech recognition or automatic speech recognition (ASR), computer speech recognition, or speech-to-text, enables a software program to process human speech into a written/text format. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Anaphora.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | In linguistics, an anaphora is a reference to a noun by way of a pronoun. For example, in the sentence, “While John didn’t like the appetizers, he enjoyed the entrée,” the word “he” is an anaphora. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/ControlNet.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | A specialized technique or technology used in machine learning to control or guide the generation process of models, often to achieve more precise or targeted outcomes.[^1] 5 | 6 | ### Footnotes 7 | 8 | [^1]: https://www.synthetic.work/news/presenting-stable-cascade/ -------------------------------------------------------------------------------- /terms/Knowledge Graphs.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Machine-readable data structures representing knowledge of the physical and digital worlds and their relationships. Knowledge graphs adhere to the graph model — a network of nodes and links. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Langops (Language Operations).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The workflows and practices that support the training, creation, testing, production deployment and ongoing curation of language models and natural language solutions. Learn more 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Language Data.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Language data is data made up of words; it is a form of unstrcutured data. This is qualitative data and also known as text data, but simply it refers to the written and spoken words in language. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Temperature.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A parameter that controls the degree of randomness or unpredictability of the LLM output. A higher value means greater deviation from the input; a lower value means the output is more deterministic. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Auto-Classification.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The application of machine learning, natural language processing (NLP), and other AI-guided techniques to automatically classify text in a faster, more cost-effective, and more accurate manner. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Context Length.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The number of tokens that the model can use at a time. The higher the context length, the more memory the model needs to train and the slower it is to run. E.g. Llama 2 can manage up to 4096 tokens. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Data Labelling.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A technique through which data is marked to make objects recognizable by machines. Information is added to various data types (text, audio, image and video) to create metadata used to train AI models. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Multimodal Models And Modalities.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Language models that are trained on and can understand multiple data types, such as words, images, audio and other formats, resulting in increased effectiveness in a wider range of tasks 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Training Set.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A training set is the pre-tagged sample data fed to an ML algorithm for it to learn about a problem, find patterns and, ultimately, produce a model that can recognize those same patterns in future analyses. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Sillytavern (Aka St).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | SillyTavern is a user interface you can install on your computer (and Android phones) that allows you to interact with text generation AIs and chat/roleplay with characters you or the community create. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Cataphora.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | In linguistics, a cataphora is a reference placed before any instance of the noun it refers to. For example, in the sentence, “Though he enjoyed the entrée, John didn’t like the appetizers,” the word “he” is a cataphora. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Post-Processing.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Procedures that can include various pruning routines, rule filtering, or even knowledge integration. All these procedures provide a kind of symbolic filter for noisy and imprecise knowledge derived by an algorithm. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Prompt.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A few words that you give to the model to start generating text. For example, if you want to generate a poem, you can give the model the first line of the poem as a prompt. The model will then generate the rest of the poem! 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Categorization.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Categorization is a natural language processing function that assigns a category to a document. Want to get more about categorization? Read our blog post “ How to Remove Pigeonholing from Your Classification Process“. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Content Enrichment Or Enrichment.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The process of applying advanced techniques such as machine learning, artificial intelligence, and language processing to automatically extract meaningful information from your text-based documents. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Knowledge Engineering.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A method for helping computers replicate human-like knowledge. Knowledge engineers build logic into knowledge-based systems by acquiring, modeling and integrating general or domain-specific knowledge into a model. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Thesauri.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Language or terminological resource “dictionary” describing relationships between lexical words and phrases in a formalized form of natural language(s), enabling the use of descriptions and relationships in text processing. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Moe Merging.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Experimental branch in mergekit that allows building a MoE-like model combining different models. You specify which models and which types of prompts you want each expert to handle, hence ending with expert task-specialization. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Morphological Analysis.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Breaking a problem with many known solutions down into its most basic elements or forms, in order to more completely understand them. Morphological analysis is used in general problem solving, linguistics and biology. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Nlt (Aka Natural Language Technology).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A subfield of linguistics, computer science and artificial intelligence (AI) dealing with Natural Language Processing (NLP), Natural Language Undestanding (NLU), and Natural Language Generation (NLG). 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Pemt (Aka Post Edit Machine Translation).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Solution allows a translator to edit a document that has already been machine translated. Typically, this is done sentence-by-sentence using a specialized computer-assisted-translation application. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Rules-Based Machine Translation (Rbmt).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Considered the “Classical Approach” of machine translation it is based on linguistic information about source and target that allow words to have different meaning depending on the context. Sampling: 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Entity.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | An entity is any noun, word or phrase in a document that refers to a concept, person, object, abstract or otherwise (e.g., car, Microsoft, New York City). Measurable elements are also included in this group (e.g., 200 pounds, 14 fl. oz.) 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Category Trees.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Enables you to view all of the rule-based categories in a collection. Used to create categories, delete categories, and edit the rules that associate documents with categories. Is also called a taxonomy, and is arranged in a hierarchy. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Conversational Models.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The LLM Leaderboard should be mostly to compare base models, not as much for conversational models. It still provides some useful signal about the conversational models, but this should not be the final way to evaluate them. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Environmental, Social, And Governance (Esg).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | An acronym initially used in business and government pertaining to enterprises’ societal impact and accountability; reporting in this area is governed by a set of binding and voluntary regulatory reporting. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Rl.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Reinforcement learning is a type of machine learning that uses rewards to train a model. For example, you can train a model to play a game by giving it a reward when it wins and a punishment when it loses. The model will learn to win the game! 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Specialized Corpora.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A focused collection of information or training data used to train an AI. Specialized corpora focuses on an industry — for example, banking, Insurance or health — or on a specific business or use case, such as legal documents. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Uncensored Models.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Many models have strong alignment that prevents doing things such as asking Llama to kill a Linux process. Training uncensored models aims to remove specific biases engrained in the decision-making process of fine-tuning a model. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Accuracy.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Accuracy is a scoring system in binary classification (i.e., determining if an answer or output is correct or not) and is calculated as (True Positives + True Negatives) / (True Positives + True Negatives + False Positives + False Negatives). 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Llama 2.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | An open-access pre-trained model released by Meta. It led to another explosion of very cool projects, and this one was not leaked! The license is not technically open-source but it's still quite open and permissive, even for commercial use cases. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Emotion Ai (Aka Affective Computing).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | AI to analyze the emotional state of a user (via computer vision, audio/voice input, sensors and/or software logic). It can initiate responses by performing specific, personalized actions to fit the mood of the customer. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Explainable Aiexplainability.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | An AI approach where the performance of its algorithms can be trusted and easily understood by humans. Unlike black-box AI, the approach arrives at a decision and the logic can be seen behind its reasoning and results. Learn more 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Hugging Face (Hf).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | HuggingFace is the standard website for distributing these AI weights openly; essentially all releases for local LLMs, whether those are finetunes or fully pretrained models from scratch, are hosted on this website in some form or another. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Test Set.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A test set is a collection of sample documents representative of the challenges and types of content an ML solution will face once in production. A test set is used to measure the accuracy of an ML system after it has gone through a round of training. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Composite Ai.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The combined application of different AI techniques to improve the efficiency of learning in order to broaden the level of knowledge representations and, ultimately, to solve a wider range of business problems in a more efficient manner. Learn more 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Lora.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | One of the most popular PEFT techniques. It adds low-rank "update matrices." The base model is frozen and only the update matrices are trained. This can be used for image classification, teaching Stable Diffusion the concept of your pet, or LLM fine-tuning. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Unstructured Data.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Unstructured data do not conform to a data model and have no rigid structure. Lacking rigid constructs, unstructured data are often more representative of “real world” business information (examples – Web pages, images, videos, documents, audio). 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Knowledge Model.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A process of creating a computer interpretable model of knowledge or standards about a language, domain, or process(es). It is expressed in a data structure that enables the knowledge to be stored in a database and be interpreted by software. Learn more 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Pretraining.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The first step in training a foundation model, usually done as an unsupervised learning phase. Once foundation models are pretrained, they have a general capability. However, foundation models need to be improved through fine-tuning to gain greater accuracy. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Classification.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Techniques that assign a set of predefined categories to open-ended text to be used to organize, structure, and categorize any kind of text – from documents, medical records, emails, files, within any application and across the web or social media networks. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Cognitive Map.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A mental representation (otherwise known as a mental palace) which serves an individual to acquire, code, store, recall, and decode information about the relative locations and attributes of phenomena in their environment. Command-R+ (aka CR+ Command R Plus): 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Pretrained Model.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A model trained to accomplish a task — typically one that is relevant to multiple organizations or contexts. Also, a pretrained model can be used as a starting point to create a fine-tuned contextualized version of a model, thus applying transfer learning. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Linked Data.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Linked data is an expression that informs whether a recognizable store of knowledge is connected to another one. This is typically used as a standard reference. For instance, a knowledge graph in which every concept/node is linked to its respective page on Wikipedia. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Model.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A machine learning model is the artifact produced after an ML algorithm has processed the sample data it was fed during the training phase. The model is then used by the algorithm in production to analyze text (in the case of NLP) and return information and/or predictions. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Speech Analytics.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The process of analyzing recordings or live calls with speech recognition software to find useful information and provide quality assurance. Speech analytics software identifies words and analyzes audio patterns to detect emotions and stress in a speaker’s voice. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Computational Linguistics (Text Analytics, Text Mining).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Computational linguistics is an interdisciplinary field concerned with the computational modeling of natural language. Find out more about Computational linguistics on our blog reading this post “ Why you need text analytics“. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Corpus.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The entire set of language data to be analyzed. More specifically, a corpus is a balanced collection of documents that should be representative of the documents an NLP solution will face in production, both in terms of content as well as distribution of topics and concepts. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Structured Data.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Structured data is the data which conforms to a specific data model, has a well-defined structure, follows a consistent order and can be easily accessed and used by a person or a computer program. Structured data are usually stored in rigid schemas such as databases. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Grounding.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The ability of generative applications to map the factual information contained in a generative output or completion. It links generative applications to available factual sources — for example, documents or knowledge bases — as a direct citation, or it searches for new links. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Nlg (Aka Natural Language Generation).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Solutions that automatically convert structured data, such as that found in a database, an application or a live feed, into a text-based narrative. This makes the data easier for users to access by reading or listening, and therefore to comprehend. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Semantic Search.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The use of natural language technologies to improve user search capabilities by processing the relationship and underlying intent between words by identifying concepts and entities such as people and organizations are revealed along with their attributes and relationships. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Botshit.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Content generated by chatbots, which is not based on truth (also known as [Hallucinations](../Hallucinations.md)), can be problematic when humans use it uncritically for communication and decision-making tasks. (Hannigan et al. 2024). [^1] 6 | 7 | **Related**: 8 | 9 | ### Citations 10 | 11 | [^1]: https://x.com/simonw/status/1788574318970229152 -------------------------------------------------------------------------------- /terms/Disambiguation.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Disambiguation, or word-sense disambiguation, is the process of removing confusion around terms that express more than one meaning and can lead to different interpretations of the same string of text. Want to learn more? Read our blog post “ Disambiguation: The Cornerstone of NLU“. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Foundational Model.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A baseline model used for a solution set, typically pretrained on large amounts of data using self-supervised learning. Applications or other models are used on top of foundational models — or in fine-tuned contextualized versions. Examples include BERT, GPT-n, Llama, DALL-E, etc. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Peft (Parameter-Efficient Fine-Tuning).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | It's a family of methods that allow fine-tuning models without modifying all the parameters. Usually, you freeze the model, add a small set of parameters, and just modify it. It hence reduces the amount of compute required and you can achieve very good results! 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Subject-Action-Object (SAO).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Subject-Action-Object (SAO) is an NLP function that identifies the logical function of portions of sentences in terms of the elements that are acting as the subject of an action, the action itself, the object receiving the action (if one exists), and any adjuncts if present. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Hybrid Ai.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Hybrid AI is any artificial intelligence technology that combines multiple AI methodologies. In NLP, this often means that a workflow will leverage both symbolic and machine learning techniques. Want to learn more about hybrd AI? Read this blog post “ What Is Hybrid Natural Language Understanding?“. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Mistral.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Mistral AI is a French company that also distributes open weight models. They are currently known for Mistral 7b and Mixtral 8x7b (which is a 47b parameters total Mixture of Experts.) Unlike the Llama series, the models they've released are licensed as Apache 2.0. Mixtral 8x22B (aka Maxtral Bistral): 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Treemap.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Treemaps display large amounts of hierarchically structured (tree-structured) data. The space in the visualization is split up into rectangles that are sized and ordered by a quantitative variable. The levels in the hierarchy of the treemap are visualized as rectangles containing other rectangles. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Computational Semantics (Semantic Technology).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Computational semantics is the study of how to automate the construction and reasoning of meaning representations of natural language expressions. Learn more about Computational semantics on our blog reading this post “ Word Meaning and Sentence Meaning in Semantics“. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Model Parameter.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | These are parameters in the model that are determined by using the training data. They are the fitted/configured variables internal to the model whose value can be estimated from data. They are required by the model when making predictions. Their values define the capability and fit of the model. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Bert (Aka Bidirectional Encoder Representation From Transformers).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Google’s technology. A large scale pretrained model that is first trained on very large amounts of unannotated data. The model is then transferred to an NLP task where it is fed another smaller task-specific dataset which is used to fine-tune the final model. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Large Language Models (Llm).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A supervised learning algorithm that uses ensemble learning method for regression. Ensemble learning method is a technique that combines predictions from multiple machine learning algorithms to make a more accurate prediction than a single model. Learn about expert.ai’s LLM trained for Insurance. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Evals.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Evals in LLM lingo refer to a framework for evaluating Large Language Models (LLMs) and LLM systems. It is an open-source framework used for assessing the performance and capabilities of these language models[^5]. 6 | 7 | Citations: 8 | [^5] https://community.openai.com/t/gpt-4-has-been-severely-downgraded-topic-curation/304946 9 | 10 | **Related**: 11 | 12 | ### Citations -------------------------------------------------------------------------------- /terms/Semantics.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Semantics is the study of the meaning of words and sentences. It concerns the relation of linguistic forms to non-linguistic concepts and mental representations to explain how sentences are understood by the speakers of a language. Learn more about semantics on our blog reading this post “ Introduction to Semantics“. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Intelligent Document Processing (Idp) Or Intelligent Document Extraction And Processing (Idep).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | This is the ability to automically read and convert unstructured and semi-structured data, identify usable data and extract it, then leveraged it via automated processes. IDP is often an enabling technology for Robotic Process Automation (RPA) tasks. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Relations.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The identification of relationships is an advanced NLP function that presents information on how elements of a statement are related to each other. For example, “John is Mary’s father” will report that John and Mary are connected, and this datapoint will carry a link property that labels the connection as “family” or “parent-child.” 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Tuning (Aka Model Tuning Or Fine Tuning).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The procedure of re-training a pre-trained language model using your own custom data. The weights of the original model are updated to account for the characteristics of the domain data and the task you are interested modeling. The customization generates the most accurate outcomes and best insights. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Logits.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The logits are the raw scores that the model creates before they are turned into probabilities, it's the final layer before you get your output. During training, all the tokens get to be a part of the end probability distribution, but the training process slowly weighs them accordingly over time to create coherent probability distributions. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Deep Learning.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Deep learning is part of a broader family of machine learning methods based on artificial neural networks with representation learning. In other words, deep learning models can learn to classify concepts from images, text or sound. In this blog post “ Word Meaning and Sentence Meaning in Semantics” you can find more about Deep Learning. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Natural Language Understanding.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A subset of natural language processing, natural language understanding is focused on the actual computer comprehension of processed and analyzed unstructured language data. This is enabled via semantics. Learn more about Natural Language Understanding (NLU) reading our blog post “What Is Natural Language Understanding?”. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Sentiment Analysis.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Sentiment analysis is an NLP function that identifies the sentiment in text. This can be applied to anything from a business document to a social media post. Sentiment is typically measured on a linear scale (negative, neutral or positive), but advanced implementations can categorize text in terms of emotions, moods, and feelings. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Conversational Ai.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Used by developers to build conversational user interfaces, chatbots and virtual assistants for a variety of use cases. They offer integration into chat interfaces such as messaging platforms, social media, SMS and websites. A conversational AI platform has a developer API so third parties can extend the platform with their own customizations. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Insight Engines.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | An insight engine, also called cognitive search or enterprise knowledge discovery. It applies relevancy methods to describe, discover, organize and analyze data. It combines search with AI capabilities to provide information for users and data for machines. The goal of an insight engine is to provide timely data that delivers actionable intelligence. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Model Drift.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Model drift is the decay of models’ predictive power as a result of the changes in real world environments. It is caused due to a variety of reasons including changes in the digital environment and ensuing changes in relationship between variables. An example is a model that detects spam based on email content and then the content used in spam was changed. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Perplexity.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Perplexity is a measurement for how predictable a specific sequence is to a language model. In the open source world, this metric is typically used to objectively compare how a model performs under different quantization conditions compared to the original model. For example, Mixtral's base model usually scores at around ~4 ppl for Wiki text style data. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Similarity (And Correlation).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Similarity is an NLP function that retrieves documents similar to a given document. It usually offers a score to indicate the closeness of each document to that used in a query. However, there are no standard ways to measure similarity. Thus, this measurement is often specific to an application versus generic or industry-wide use cases. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Text Summarization.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A range of techniques that automatically produce short textual summaries representing longer or multiple texts. The principal purpose of this technology is to reduce employee time and effort required to acquire insight from content, either by signaling the value of reading the source(s), or by delivering value directly in the form of the summary. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Generative Summarization.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Using LLM functionality to take text prompt inputs like long form chats, emails, reports, contracts, policies, etc and distilling them down to their core content, generating summaries from the text prompts for quick comprehension. Thus using pre-trained language models and context understanding to produce concise, accurate and relevant summaries. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Summarization (Text).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Text summarization is the process of creating a short, accurate, and fluent summary of a longer text document. The goal is to reduce the size of the text while preserving its important information and overall meaning. There are two main types of text summarization: Extractive Summarization Generative Summarization also know as Abstractive Summarization 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Text Analytics.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Techniques used to process large volumes of unstructured text (or text that does not have a predefined, structured format) to derive insights, patterns, and understanding; the process can include determining and classifying the subjects of texts, summarizing texts, extracting key entities from texts, and identifying the tone or sentiment of texts. Learn more 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Pre-Processing.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A step in the data mining and data analysis process that takes raw data and transforms it into a format that can be understood and analyzed by computers. Analyzing structured data, like whole numbers, dates, currency and percentages is straigntforward. However, unstructured data, in the form of text and images must first be cleaned and formatted before analysis. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/WizardLM 2.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | WizardLM-2 8x22B, WizardLM-2 70B, and WizardLM-2 7B. 6 | Method Overview: As the natural world's human-generated data becomes increasingly exhausted through LLM training, we believe that: the data carefully created by AI and the model step-by-step supervised by AI will be the sole path towards more powerful AI. 7 | WizardLM: A research team from Microsoft...but also a Discord community. 8 | 9 | ### Related Articles 10 | 11 | ### Citations 12 | -------------------------------------------------------------------------------- /terms/Alice and Bob.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | The word 'Alice' is sometimes used a synonym for 'agent' especially in the Oxford-related academic world. 'Alice' or 'Bob' are placeholder names for otherwise abstract agents or particles. [^1] [^2] 6 | 7 | **Related**: 8 | 9 | ### Citations 10 | 11 | [^1]: https://en.m.wikipedia.org/wiki/Alice_and_Bob 12 | [^2]: Floridi, Luciano. _The Ethics of Artificial Intelligence: Principles, Challenges, and Opportunities_. Oxford University Press, 2023. -------------------------------------------------------------------------------- /terms/Imatrix.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | llama.cpp also recently introduced exllama-esque quantization through the "importance matrix" calculations (otherwise known as an "imatrix".) Technically this is a distinct technique from exllamav2, but the results are of comparable quality. Before that k-quants were also introduced to improve upon the linear rounding technique, which can be used in tandem with the imatrix. Inference / Inferencing: 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Zero-Shot.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A type of prompt that is used to generate text without fine-tuning. The model is not trained on any specific task. It is only trained on a large dataset of text. For example, you can give the model the first line of a poem and ask it to generate the rest of the poem. The model will do its best to generate a poem, even though it has never seen a poem before! When you use ChatGPT, you often do zero-shot generation! 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Base Vs Conversational.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A pre-trained model is not specifically trained to "behave" in a conversational manner. If you try to use a base model (e.g. GPT-3, Mistral, Llama) directly to do conversations, it won't work as well as the fine-tuned conversational variant (ChatGPT, Mistral Instruct, Llama Chat). When looking at benchmarks, you want to compare base models with base models and conversational models with conversational models. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Ai Feedback (Aif).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | AI Feedback refers to the mechanisms and processes by which these models receive, integrate, and respond to user inputs, corrections, or other forms of feedback to improve accuracy, relevancy, and safety of their outputs. This often involves techniques such as reinforcement learning from human feedback (RLHF), where models are fine-tuned based on evaluations of their performance against human judgment or preferences. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Auto-Regressive.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A type of model that generates text one token at a time. It is auto-regressive because it uses its own predictions to generate the next token. For example, the model might receive as input "Today's weather" and generate the next token, "is". It will then use "Today's weather is" as input and generate the next token, "sunny". It will then use "Today's weather is sunny" as input and generate the next token, "and". And so on. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Natural Language Processing.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A subfield of artificial intelligence and linguistics, natural language processing is focused on the interactions between computers and human language. More specifically, it focuses on the ability of computers to read and analyze large volumes of unstructured language data (e.g., text). Read our blog post “ 6 Real-World Examples of Natural Language Processing” to learn more about Natural Language Processing (NLP). 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Question & Answer (Q&A).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | An AI technique that allows users to ask questions using common everyday language and receive the correct response back. With the advent of large language models (LLMs), question and answering has evolved to let users ask questions using common everyday language and use Retrieval Augmented Generation (RAG) approaches to generate a complete answer from the text fragments identified in the target document or corpus. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Co-Occurrence.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A co-occurrence commonly refers to the presence of different elements in the same document. It is often used in business intelligence to heuristically recognize patterns and guess associations between concepts that are not naturally connected (e.g., the name of an investor often mentioned in articles about startups successfully closing funding rounds could be interpreted as the investor is particularly good at picking his or her investments.). 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Ontology.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | An ontology is similar to a taxonomy, but it enhances its simple tree-like classification structure by adding properties to each node/element and connections between nodes that can extend to other branches. These properties are not standard, nor are they limited to a predefined set. Therefore, they must be agreed upon by the classifier and the user. Read our blog post “ Understanding Ontology and How It Adds Value to NLU” to learn more about the ontologies. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Semi-Structured Data.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Data that is stuctured in some way but does not obey the tablular structure of traditional databases or other conventional data tables most commonly organized in rows and columns. Attributes of the data are different even though they may be grouped together. A simple example is a form; a more advanced example is a object database where the data is represented in the form of objects that are related (e.g. automobile make relates to model relates to trim level). 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Pre-Training.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Training a model on a very large dataset (trillion of tokens) to learn the structure of language. Imagine you have millions of dollars, as a good GPU-Rich. You usually scrape big datasets from the internet and train your model on them. This is called pre-training. The idea is to end with a model that has a strong understanding of language. This does not require labeled data! This is done before fine-tuning. Examples of pre-trained models are GPT-3, Llama 2, and Mistral. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Quantization.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | LLMs can take gigabytes of memory to store, which limits what can be run on consumer hardware. But quantization can dramatically compress models, making a wider selection of models available to developers. You can often reduce model size by 4x or more while maintaining reasonable performance. [^1] 6 | 7 | As models get bigger and bigger, quantization becomes more important for making models practical and accessible. [^1] 8 | 9 | **Related**: 10 | 11 | ### Citations 12 | 13 | [^1]: https://x.com/AndrewYNg/status/1779905922602782752 -------------------------------------------------------------------------------- /terms/Retrieval Augmented Generation (Rag).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Retrieval-augmented generation (RAG) is an AI technique for improving the quality of LLM-generated responses by including trusted sources of knowledge, outside of the original training set, to improve the accuracy of the LLM’s output. Implementing RAG in an LLM-based question answering system has benefits: 1) assurance that an LLM has access to the most current, reliable facts, 2) reduce hallucinations rates, and 3) provide source attribution to increase user trust in the output. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Gqa.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | GQA stands for "Grouped Query Attention". This is a training strategy used to reduce the memory footprint of large Transformers by letting multiple queries share keys and values. I don't quite understand the technical details beyond that, but the important element is that it prevents the larger context size from being extremely expensive in terms of memory requirements, with almost no degradation on the end results. Llama 2 70b, Mixtral, Yi 34b, Mistral 7b are all examples of modern models that were trained with GQA. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Symbolic Methodology.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A symbolic methodology is an approach to developing AI systems for NLP based on a deterministic, conditional approach. In other words, a symbolic approach designs a system using very specific, narrow instructions that guarantee the recognition of a linguistic pattern. Rule-based solutions tend to have a high degree of precision, though they may require more work than ML-based solutions to cover the entire scope of a problem, depending on the application. Want to learn more about symbolic methodology?. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Total Parameters.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Approximately 47 billion, but only 12 billion are actively used at any one time. 6 | Configuration: Consists of eight experts, each with around 7 billion parameters. 7 | Performance: Demonstrates superior performance in most real-world tasks compared to GPT-3.5 and the former leading open-source LLM, Llama 2 70b. 8 | Mixture of Experts (MoE): A neural network architecture where specific layers are replaced with multiple smaller networks, or "experts," managed by a gate network or router. Key points include: 9 | 10 | ### Related Articles 11 | 12 | ### Citations 13 | -------------------------------------------------------------------------------- /terms/Tokenizer.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Before language models are trained, the data used to create them gets split into pieces with a "dictionary" of sorts, and each piece of this dictionary represents a different word (or a part of a word). This is so they can meaningfully learn patterns from the data. The "words" in this dictionary are referred to as tokens, and the "dictionary" is called a Tokenizer. 6 | Tokens: A unit of content corresponding to a subset of a word. Tokens are processed internally by LLMs and can also be used as metrics for usage and billing. 7 | 8 | ### Related Articles 9 | 10 | ### Citations 11 | -------------------------------------------------------------------------------- /terms/Part-Of-Speech Tagging.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A Part-of-Speech (POS) tagger is an NLP function that identifies grammatical information about the elements of a sentence. Basic POS tagging can be limited to labeling every word by grammar type, while more complex implementations can group phrases and other elements in a clause, recognize different types of clauses, build a dependency tree of a sentence, and even assign a logical function to every word (e.g., subject, predicate, temporal adjunct, etc.). Find out more about Part-of-Speech (POS) tagger in this article on our Community. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Responsible Ai.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Responsible AI is a broad term that encompasses the business and ethical choices associated with how organizations adopt and deploy AI capabilities. Generally, Responsible AI looks to ensure Transparent (Can you see how an AI model works?); Explainable (Can you explain why a specific decision in an AI model was made?); Fair (Can you ensure that a specific group is not disadvantaged based on an AI model decision?); and Sustainable (Can the development and curation of AI models be done on an environmentally sustainable basis?) use of AI. Learn more 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Rlhf (Reinforcement Learning With Human Feedback).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A type of fine-tuning that uses reinforcement learning (RL) and human-generated feedback. Thanks to the introduction of human feedback, the end model ends up being very good for things such as conversations! It kicks off with a base model that generates bunch of conversations. Humans then rate the answers (preferences). The preferences are used to train a Reward Model that generates a score for a given text. Using Reinforcement Learning, the initial LM is trained to maximize the score generated by the Reward Model. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Controlled Vocabulary.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A controlled vocabulary is a curated collection of words and phrases that are relevant to an application or a specific industry. These elements can come with additional properties that indicate both how they behave in common language and what meaning they carry, in terms of topic and more. While the value of a controlled vocabulary is similar to that of taxonomy, they differ in that the nodes in taxonomy are only labels representing a category, while the nodes in a controlled vocabulary represent the words and phrases that must be found in a text. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Definition.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Parameters are numerical values adjusted during pre-training and fine-tuning phases. 6 | Scale Indication: Represented by figures like 7b, 13b, and 34b, indicating billions of parameters. 7 | Importance: While a higher parameter count typically suggests better model performance due to finer text pattern representation, it is not the sole factor; the quality and quantity of training data also significantly influence effectiveness. 8 | ollama: An open-source tool to run LLMs locally. There are multiple web/desktop apps and terminal integrations on top of it. 9 | 10 | ### Related Articles 11 | 12 | ### Citations 13 | -------------------------------------------------------------------------------- /terms/Finetunes Fine-Tuning.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Training a model on a small (labeled) dataset to learn a specific task. This is done after pre-training. Imagine you have a few dollars, as a good fellow GPU-Poor. Rather than training a model from scratch, you pick a pre-trained (base) model and fine-tune it. You usually pick a small dataset of few hundreds-thousands of samples. You then pass it to the model and train it on it. This is called fine-tuning. The idea is to end with a model that has a strong understanding of a specific task. Examples of fine-tuned models are ChatGPT, Vicuna, and Mistral Instruct. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Cosine Similarity.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Cosine similarity is a metric used to measure the similarity of two vectors. Specifically, it measures the similarity in the direction or orientation of the vectors ignoring differences in their magnitude or scale. Both vectors need to be part of the same inner product space, meaning they must produce a scalar through inner product multiplication. The similarity of two vectors is measured by the cosine of the angle between them.[^1] 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | 11 | [^1]: https://www.learndatasci.com/glossary/cosine-similarity/#:~:text=Cosine%20Similarity%20is%20widely%20used,or%20users%20in%20recommendation%20systems -------------------------------------------------------------------------------- /terms/Recall.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Given a set of results from a processed document, recall is the percentage value that indicates how many correct results have been retrieved based on the expectations of the application. It can apply to any class of a predictive AI system such as search, categorization and entity recognition. For example, say you have an application that is supposed to find all the dog breeds in a document. If the application analyzes a document that mentions 10 dog breeds but only returns five values (all of which are correct), the system will have performed at 50% recall. Find out more about recall on our Community reading this article. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Llama.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A pre-trained model trained by Meta, shared with some groups in a private access, and then leaked. It led to an explosion of cool projects. 6 | 7 | A language model series created by Meta. Llama 1 was originally leaked in February 2023; Llama 2 then officially released later that year with openly available model weights & a permissive license. Kicked off the initial wave of open source developments that have been made when it comes to open source language modeling. The Llama series comes in four distinct sizes: 7b, 13b, 34b (only Code Llama was released for Llama 2 34b), and 70b. As of writing, the hotly anticipated Llama 3 has yet to arrive. 8 | ### Related Articles 9 | 10 | ### Citations 11 | -------------------------------------------------------------------------------- /terms/Llama 3.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | An open-source large language model (LLM) developed by Meta, designed to advance AI capabilities through enhanced scalability and performance. It features multiple versions with different parameter sizes, such as 8 billion (8B) and 70 billion (70B), and is trained on over 15 trillion (15T) tokens, including multilingual and code data. Llama 3 incorporates advanced technologies like grouped query attention and extensive instruction tuning to improve efficiency and effectiveness in various AI tasks. It is part of Meta's initiative to democratize AI, making powerful tools accessible for broader innovation and responsible application in technology. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Inference Inferencing.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | "Inference" is the term for actually using the model to make predictions. When people discuss inference speed, they're usually concerned about two things: The prompt processing speed, and the generation speed. 6 | Both of these can be measured in "tokens per second", but the numbers for each tend to be different due to how batching works (naturally, it's a lot faster to evaluate 500 tokens at once compared to evaluating 1 token at a time, which is what is happening during the generation process). 7 | Inference Engine: A component of a [expert] system that applies logical rules to the knowledge base to deduce new or additional information. 8 | 9 | ### Related Articles 10 | 11 | ### Citations 12 | -------------------------------------------------------------------------------- /terms/Precision.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Given a set of results from a processed document, precision is the percentage value that indicates how many of those results are correct based on the expectations of a certain application. It can apply to any class of a predictive AI system such as search, categorization and entity recognition. For example, say you have an application that is supposed to find all the dog breeds in a document. If the application analyzes a document that mentions 10 dog breeds but only returns five values (all of which are correct), the system will have performed at 100% precision. Even if half of the instances of dog breeds were missed, the ones that were returned were correct. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Machine Learning (Ml).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | Machine learning is the study of computer algorithms that can improve automatically through experience and the use of data. It is seen as a part of artificial intelligence. Machine learning algorithms build a model based on sample data, known as “training data,” in order to make predictions or decisions without being explicitly programmed to do so. In NLP, ML-based solutions can quickly cover the entire scope of a problem (or, at least of a corpus used as sample data), but are demanding in terms of the work required to achieve production-grade accuracy. Want to get more about machine learning? Read this post “ What Is Machine Learning? A Definition” on our blog. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Taxonomy.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A taxonomy is a predetermined group of classes of a subset of knowledge (e.g., animals, drugs, etc.). It includes dependencies between elements in a “part of” or “type of” relationship, giving itself a multi-level, tree-like structure made of branches (the final node or element of every branch is known as a leaf). This creates order and hierarchy among knowledge subsets. Companies use taxonomies to more concisely organize their documents which, in turn, enables internal or external users to more easily search for and locate the documents they need. They can be specific to a single company or become de-facto languages shared by companies across specific industries. Find out more about taxonomy reading our blog post “ What Are Taxonomies and How Should You Use Them?“. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/F-Score (F-Measure, F1 Measure).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | An F-score is the harmonic mean of a system’s precision and recall values. It can be calculated by the following formula: 2 x [(Precision x Recall) / (Precision + Recall)]. Criticism around the use of F-score values to determine the quality of a predictive system is based on the fact that a moderately high F-score can be the result of an imbalance between precision and recall and, therefore, not tell the whole story. On the other hand, systems at a high level of accuracy struggle to improve precision or recall without negatively impacting the other. Critical (risk) applications that value information retrieval more than accuracy (i.e., producing a large number of false positives but virtually guaranteeing that all the true positives are found) can adopt a different scoring system called F2 measure, where recall is weighed more heavily. The opposite (precision is weighed more heavily) is achieved by using the F0.5 measure. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Knowledge Graph.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | A knowledge graph is a graph of concepts whose value resides in its ability to meaningfully represent a portion of reality, specialized or otherwise. Every concept is linked to at least one other concept, and the quality of this connection can belong to different classes (see: taxonomies). The interpretation of every concept is represented by its links. Consequently, every node is the concept it represents only based on its position in the graph (e.g., the concept of an apple, the fruit, is a node whose parents are “apple tree”, “fruit”, etc.). Advanced knowledge graphs can have many properties attached to a node including the words used in language to represent a concept (e.g., “apple” for the concept of an apple), if it carries a particular sentiment in a culture (“bad”, “beautiful”) and how it behaves in a sentence. Learn more about knowledge graph reafding this blog post “ Knowledge Graph: The Brains Behind Symbolic AI” on our blog. 6 | 7 | ### Related Articles 8 | 9 | ### Citations 10 | -------------------------------------------------------------------------------- /terms/Agentic Al Design Patterns.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | [Agentic Workflows](./Agentic%20Workflows.md), like in programming can abide to different design patterns. 5 | 6 | Here are some popular design patterns for agentic workflows: 7 | 8 | - [Reflection](https://x.com/AndrewYNg/status/1773393357022298617): The LLM examines its own work to come up with ways to improve it. 9 | - [Tool use](https://x.com/AndrewYNg/status/1775951610059141147): The LLM is given tools such as web search, code execution, or any other function to help it gather information, take action, or process data. 10 | - [Planning](https://x.com/AndrewYNg/status/1779606380665803144): The LLM comes up with, and executes, a multistep plan to achieve a goal (for example, writing an outline for an essay, then doing online research, then writing a draft, and so on). 11 | - [Multi-agent collaboration](https://twitter.com/AndrewYNg/status/1780991671855161506): More than one AI agent work together, splitting up tasks and discussing and debating ideas, to come up with better solutions than a single agent would. [^1] 12 | 13 | ### Citations 14 | 15 | [^1]: https://www.deeplearning.ai/the-batch/issue-245/ -------------------------------------------------------------------------------- /terms/Model Reliability.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | The degree to which an artificial intelligence model consistently produces accurate, dependable, and repeatable outputs or results. [^1] 5 | 6 | System performance, reliability, and user confidence in AI model output is affected as much by the quality of the model design as the data going into it. [^2] 7 | 8 | To achieve AI that is robust and reliable, companies need to ensure their AI algorithms produce the right results for each new data set. They also need established processes for handling issues and inconsistencies if and when they arise. The human factor is a critical element here: understanding how human input affects reliability; determining who the right people are to provide input; and ensuring those people are properly equipped and trained—particularly with regard to bias and ethics. [^3] 9 | 10 | ### Footnotes 11 | 12 | [^1]: https://www.synthetic.work/news/openai-discloses-its-number-of-chatgpt-for-enterprise-customers-to-date/ 13 | [^2]: https://www.technologyreview.com/2020/07/06/1004823/beyond-the-ai-hype-cycle-trust-and-the-future-of-ai/ 14 | [^3]: https://www.technologyreview.com/2020/03/25/950291/trustworthy-ai-is-a-framework-to-help-manage-unique-risk/ -------------------------------------------------------------------------------- /terms/Effective Accelerationism.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | Effective accelerationism, often abbreviated as "e/acc", is a 21st-century philosophical movement that advocates for an explicitly pro-technology stance. Its proponents believe that unrestricted technological progress (especially driven by artificial intelligence) is a solution to universal human problems like poverty, war and climate change. They see themselves as a counterweight to more cautious views on technological innovation, often giving their opponents the derogatory labels of "doomers" or "decels" (short for deceleration). [^1] 5 | 6 | Advocates of effective accelerationism promote a perspective that not only should the advancement of artificial intelligence be permitted to continue without restrictions, but it should also be expedited. [^2] 7 | 8 | Leading the charge of e/acc is [Marc Andreessen](../Marc%20Andreessen.md), co-founder of Andreessen Horowitz, a venture-capital firm. [^2] 9 | 10 | ### Footnotes 11 | 12 | [^1]: https://en.m.wikipedia.org/wiki/Effective_accelerationism 13 | [^2]: https://www.economist.com/business/2023/11/19/the-sam-altman-drama-points-to-a-deeper-split-in-the-tech-world 14 | https://www.theinformation.com/articles/its-a-cult-inside-effective-accelerationism-the-pro-ai-movement-taking-over-silicon-valley?rc=yfiisw -------------------------------------------------------------------------------- /terms/Effective Altruism.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | “[Effective altruism](https://www.economist.com/1843/2022/11/15/the-good-delusion-has-effective-altruism-broken-bad)”, a movement that is concerned by the possibility of ai wiping out all of humanity. [^1] 5 | 6 | Both are pillars of “longtermism,” a growing strain of the ideology known as effective altruism (or EA), which is popular among an elite slice of people in tech and [politics](https://www.nytimes.com/2022/09/08/style/david-shor-democrats.html). [^2] 7 | 8 | Since its birth in the late 2000s, effective altruism has aimed to answer the question “How can those with means have the _most_ impact on the world in a quantifiable way?”—and supplied clear methodologies for calculating the answer.[^2] 9 | 10 | “[Longtermism](https://aeon.co/essays/why-longtermism-is-the-worlds-most-dangerous-secular-credo),” the belief that unlikely but existential threats like a humanity-destroying AI revolt or international biological warfare are humanity’s most pressing problems, is integral to EA today. [^2] 11 | 12 | ### Footnotes 13 | 14 | [^1]: https://www.economist.com/business/2023/11/19/the-sam-altman-drama-points-to-a-deeper-split-in-the-tech-world 15 | [^2]: https://www.technologyreview.com/2022/10/17/1060967/effective-altruism-growth/?truid=*%7CLINKID%7C*&utm_source=the_download&utm_medium=email&utm_campaign=the_download.unpaid.engagement&utm_term=*%7CSUBCLASS%7C*&utm_content=*%7CDATE:m-d-Y%7C* -------------------------------------------------------------------------------- /terms/Red-Teaming.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | Similar to a [penetration test](https://en.m.wikipedia.org/wiki/Penetration_test), Red Teaming is an adversarial examination of the AI system to ensure that AI systems are ethical and safe. 5 | 6 | Red teams conduct qualitative probes and adversarial tests on new models. Their goal is to break these models and pinpoint their weaknesses.[^1] 7 | 8 | In the year 2023, [OpenAI](https://en.m.wikipedia.org/wiki/OpenAI) employed 50 academics and experts to examine the capabilities of the GPT-4 model, which currently powers the premium version of ChatGPT. This process is now known as "red-teaming". [^3] 9 | 10 | Over a span of six months, this team, consisting of experts from various fields such as chemistry, nuclear weapons, law, education, and misinformation, was tasked to qualitatively probe and adversarially test the new model in an effort to identify its weaknesses. Red-teaming is a strategy also used by other organisations like Google DeepMind and Anthropic to identify and rectify the weaknesses in their software. [^2] 11 | 12 | While [RLHF](./RLHF%20(Reinforcement%20Learning%20From%20Human%20Feedback).md) and red-teaming are crucial for AI safety, they do not completely eliminate the issue of harmful AI outputs [^2] 13 | 14 | ### Footnotes 15 | 16 | [^1]: https://www.ft.com/content/0876687a-f8b7-4b39-b513-5fee942831e8 17 | [^2]: https://www.ft.com/content/f23e59a2-4cad-43a5-aed9-3aea6426d0f2 18 | [^3]: https://www.ft.com/content/f23e59a2-4cad-43a5-aed9-3aea6426d0f2 -------------------------------------------------------------------------------- /terms/Alignment Training.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | There are a few types of alignment training. The most relevant looks rather old-fashioned: Humans interact with the LLM and rate its responses good or bad, which coaxes it toward certain behaviors (such as being friendly or polite) and away from others (like profanity and abusive language). Tramèr told me that this seems to steer LLMs away from quoting their training data. He was part of a team that managed to break ChatGPT’s alignment training while studying its ability to memorize text, but he said that it works “remarkably well” in normal interactions. Nevertheless, he said, “alignment alone is not going to completely get rid of this problem.” [^1] 5 | 6 | Another potential solution is [retrieval-augmented generation](https://www.theatlantic.com/technology/archive/2023/11/google-generative-ai-search-featured-results/675899/). RAG is a system for finding answers to questions in external sources, rather than within a language model. A RAG-enabled chatbot can respond to a question by retrieving relevant webpages, summarizing their contents, and providing links. Google Bard, for example, offers a list of “additional resources” at the end of its answers to some questions. RAG isn’t bulletproof, but it [reduces the chance](https://research.ibm.com/blog/retrieval-augmented-generation-RAG) of an LLM giving incorrect information (or “hallucinating”), and it has the added benefit of avoiding copyright infringement, because sources are cited. [^1] 7 | 8 | ### Footnotes 9 | 10 | [^1]: https://www.theatlantic.com/technology/archive/2024/01/chatgpt-memorization-lawsuit/677099/ -------------------------------------------------------------------------------- /terms/Indexer (LLM).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | An indexer in the context of LLMs (Large Language Models) refers to tools that help manage and organise data for efficient querying and retrieval by LLMs. Two such tools are LlamaIndex and LangChain. 6 | 7 | [LlamaIndex](../LlamaIndex.md), previously known as GPT Index, is a data framework specifically designed for LLM apps. Its primary focus is on ingesting, structuring, and accessing private or domain-specific data. It offers a set of tools that facilitate the integration of custom data into LLMs. LlamaIndex is optimized for index querying and supports vector embeddings, making it an ideal solution for applications requiring indexing and retrieval capabilities[^2]. 8 | 9 | [LangChain](../LangChain.md) is a more general-purpose framework that can be used to build a wide variety of applications. It provides tools for loading, processing, and indexing data, as well as for interacting with LLMs. Langchain is also more flexible than LlamaIndex, allowing users to work with different types of data and LLMs[^2]. 10 | 11 | Both LlamaIndex and LangChain can be used together to enhance RAG (Retrieval-Augmented Generation) applications. LlamaIndex is particularly suitable for indexing and retrieval, while LangChain provides a more general-purpose framework for building applications with LLMs[^2]. 12 | 13 | ### Related Articles 14 | - [RAG (Retrieval-Augmented Generation)](../RAG%20(Retrieval-Augmented%20Generation).md) 15 | 16 | ### Footnotes 17 | 18 | [^1] https://alphasec.io/query-your-own-documents-with-llamaindex-and-langchain/ 19 | [^2] https://stackoverflow.com/questions/76990736/differences-between-langchain-llamaindex 20 | [^3] https://medium.aiplanet.com/implement-rag-with-knowledge-graph-and-llama-index-6a3370e93cdd?gi=ab4c15079dd7 21 | [^4] https://www.llamaindex.ai/blog/a-new-document-summary-index-for-llm-powered-qa-systems-9a32ece2f9ec 22 | [^5] https://pypi.org/project/llama-index/ -------------------------------------------------------------------------------- /terms/AI-Washing.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | After being warned about “[greenwashing](https://en.m.wikipedia.org/wiki/Greenwashing)”, companies are now being told not to engage in **ai-washing**, or claiming a product or service has been created with artificial intelligence when it has not. 5 | 6 | Examples of companies that have failed to mention humans pulling the levers behind supposedly cutting-edge AI technology. [^1] To name just a few: 7 | - Amazon's AI-based 'just walk out' checkout tech [was powered by 1,000 Indian workers](https://www.bloomberg.com/opinion/articles/2024-04-03/the-humans-behind-amazon-s-just-walk-out-technology-are-all-over-ai) manually. 8 | - Facebook [famously shut down](https://www.theverge.com/2018/1/8/16856654/facebook-m-shutdown-bots-ai) its text-based virtual assistant M in 2018 after more than two years, during which the company used human workers to train (and operate) its underlying artificial intelligence system. 9 | - A startup called x.ai, which marketed an “AI personal assistant” that scheduled meetings, [had humans doing that work instead](https://www.bloomberg.com/news/articles/2016-04-18/the-humans-hiding-behind-the-chatbots) and shut down in 2021 after it struggled to get to a point where the algorithms could work independently. 10 | - A British startup called Builder.ai sold AI software that could build apps even though it [partly relied on](https://www.wsj.com/articles/ai-startup-boom-raises-questions-of-exaggerated-tech-savvy-11565775004) software developers in India and elsewhere to do that work, according to a _Wall Street Journal_ report. 11 | 12 | [Pseudo AI](https://www.forbes.com/sites/cognitiveworld/2020/04/04/artificial-or-human-intelligence-companies-faking-ai/) or “AI washing” was widespread even before the recent generative AI boom. 13 | 14 | This "AI washing" threatens to overinflated expectations for the technology, undermining public trust and potentially setting up the booming field for a backlash. 15 | 16 | ### Footnotes 17 | 18 | [^1]: https://www.axios.com/2019/11/16/ai-washing-hidden-people -------------------------------------------------------------------------------- /terms/Trustworthy AI.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | 5 | ‘Trustworthy AI’ is a framework developed by [Deloitte](https://en.m.wikipedia.org/wiki/Deloitte) to help manage AI risk. [^1] 6 | 7 | Deloitte’s Trustworthy AI framework introduces six key dimensions that, when considered collectively in the design, development, deployment, and operational phases of AI system implementation, can help safeguard ethics and build a trustworthy AI strategy. [^1] 8 | 9 | The Trustworthy AI framework is designed to help companies identify and mitigate potential risks related to AI ethics at every stage of the AI lifecycle. Here’s a closer look at each of the framework’s six dimensions. [^1] 10 | 11 | 12 |  13 | 14 | 15 | 1. **Fair, not biased** - How AI makes decisions must be open to inspection and fully explainable 16 | For AI to be trustworthy, all participants have a right to understand how their data is being used and how the AI is making decisions. 17 | 2. **Transparent and explainable** - Clear policies must establish accountability for AI output 18 | Trustworthy AI systems need to include policies that clearly establish who is responsible and accountable for their output. 19 | 3. **Responsible and accountable** - Clear policies must establish accountability for AI output 20 | Trustworthy AI systems need to include policies that clearly establish who is responsible and accountable for their output. 21 | 4. **Robust and reliable** - AI must be as reliable as traditional systems and consistent in all conditions 22 | In order for AI to achieve widespread adoption, it must be at least as robust and reliable as the traditional systems, processes, and people it is augmenting or replacing. 23 | 5. **Respectful of privacy** - AI must comply with regulations and only use data as agreed 24 | Privacy is a critical issue for all types of data systems, but it is especially critical for AI since the sophisticated insights generated by AI systems often stem from data that is more detailed and personal. 25 | 6. **Safe and secure** - AI must be protected from risks like hacking that could harm people or data 26 | To be trustworthy, AI must be protected from cybersecurity risks that might lead to physical and/or digital harm. 27 | 28 | ### Footnotes 29 | 30 | [^1]: https://www.technologyreview.com/2020/03/25/950291/trustworthy-ai-is-a-framework-to-help-manage-unique-risk/ -------------------------------------------------------------------------------- /terms/Agentic Workflows.md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | Agentic workflows refer to a new way of interacting with large language models (LLMs) where multiple instances of LLM performs actions in an iterative way, rather than just responding to a single prompt. 5 | 6 | Agents can be customized and augmented using prompt engineering techniques and external tools that enable them to retrieve information or execute code. [^2] 7 | 8 | Multi-agent applications can be fully autonomous or moderated through “human proxy agents,” which allow users to step into the conversation between the AI agents, acting as another voice to provide oversight and control over their process. In a way, the human user is turned into a team leader overseeing a team of multiple AIs. [^2] 9 | 10 | > "Today, we mostly use LLMs in zero-shot mode, prompting a model to generate final output token by token without revising its work. This is akin to asking someone to compose an essay from start to finish, typing straight through with no backspacing allowed, and expecting a high-quality result. Despite the difficulty, LLMs do amazingly well at this task!" 11 | > "With an agentic workflow, however, we can ask the LLM to iterate over a document many times. For example, it might take a sequence of steps such as: 12 | > - Plan an outline. 13 | > - Decide what, if any, web searches are needed to gather more information. 14 | > - Write a first draft. 15 | > - Read over the first draft to spot unjustified arguments or extraneous information. 16 | > - Revise the draft taking into account any weaknesses spotted. 17 | > - And so on. 18 | > This iterative process is critical for most human writers to write good text. With AI, such an iterative workflow yields much better results than writing in a single pass." 19 | > — Andrew Ng, Co-founder of Google Brain and Former Chief Scientist at Baidu [^2] 20 | 21 | Agentic workflows, like in programming can abide to different [Agentic Al Design Patterns](./Agentic%2520Al%2520Design%2520Patterns.md#). 22 | ### Agentic Workflows Tools 23 | - Programming frameworks for agentic AI. 24 | - [Microsoft Autogen](https://github.com/microsoft/autogen) 25 | - [CrewAI](https://github.com/joaomdmoura/crewai) 26 | - [ChatDev](https://github.com/OpenBMB/ChatDev) 27 | ### Footnotes 28 | 29 | https://www.youtube.com/embed/sal78ACtGTc?si=H4mU6MDkj6tAGspu 30 | 31 | [^1]: https://twitter.com/AndrewYNg/status/1770897666702233815 32 | [^2]: https://venturebeat.com/ai/microsofts-autogen-framework-allows-multiple-ai-agents-to-talk-to-each-other-and-complete-your-tasks/ -------------------------------------------------------------------------------- /terms/RLHF (Reinforcement Learning From Human Feedback).md: -------------------------------------------------------------------------------- 1 | --- 2 | share: true 3 | --- 4 | RLHF is a method used to reduce bias in large language models. It enables the AI to learn from human preferences. It works by allowing human testers to guide the AI model towards producing the type of text output they indicate they prefer. The use of the ratings given by human testers guides the improvement of AI models. [^3][^5][^6]. 5 | 6 | ChatGPT was trained using a RLHF. It is often said that RLHF is the secret ingredient of ChatGPT. The basic idea is to take a large language model with a tendency to spit out anything it wants—in this case, GPT-3.5—and tune it by teaching it what kinds of responses human users actually prefer. [^1] 7 | 8 | OpenAI reduced the amount of misinformation and offensive text that GPT-3 produced by using reinforcement learning to train a version of the model on the preferences of human testers. [^2] 9 | 10 | [OpenAI](https://openai.com/research/instruction-following), [Google](https://www.deepmind.com/blog/building-safer-dialogue-agents), [Anthropic](https://www-files.anthropic.com/production/images/Model-Card-Claude-2.pdf), and other companies all use the technique. After a chatbot has processed massive amounts of text, human feedback helps fine-tune it. [^5] 11 | 12 | To apply RLHF, companies hire large teams of contractors to look at the responses of their AI models and rate them as “good” or “bad”. By analysing enough responses, the model becomes attuned to those judgments, and filters its responses accordingly. This basic process works to refine an AI’s responses at a superficial level. But the method is primitive, according to [Dario Amodei](../Dario%20Amodei.md), who helped develop it while previously working at OpenAI. “It’s . . . not very accurate or targeted, you don’t know why you’re getting the responses you’re getting [and] there’s lots of noise in that process,” he said. [^6] 13 | 14 | ### Footnotes 15 | 16 | [^1]: https://www.technologyreview.com/2023/03/03/1069311/inside-story-oral-history-how-chatgpt-built-openai/ 17 | [^2]: https://www.technologyreview.com/2023/02/08/1068068/chatgpt-is-everywhere-heres-where-it-came-from/ 18 | [^3]: https://www.technologyreview.com/2023/12/19/1084505/generative-ai-artificial-intelligence-bias-jobs-copyright-misinformation/ 19 | [^4]: https://www.technologyreview.com/2023/03/20/1070067/language-models-may-be-able-to-self-correct-biases-if-you-ask-them-to/ 20 | [^5]: https://www.theatlantic.com/technology/archive/2023/07/ai-chatbot-human-evaluator-feedback/674805/ 21 | [^6]: https://www.ft.com/content/f23e59a2-4cad-43a5-aed9-3aea6426d0f2 -------------------------------------------------------------------------------- /CONTRIBUTING.md: -------------------------------------------------------------------------------- 1 | # Origin 2 | 3 | The A to Z of AI was created by Alex Fazio. It is operated as a Working Group under the auspices of AI Garden and is governed by AI Garden's Committee. 4 | 5 | # Contributing 6 | 7 | Everybody is invited to suggest changes, additions and improvements to The A to Z of AI . We employ a community-driven process to develop and improve upon this shared lexicon. This glossary provides a vendor-neutral platform on which to organize a shared vocabulary around AI and automation. Contributions are welcome from all participants who abide by the project's purpose and charter. 8 | 9 | Anyone who wishes to make a contribution, may submit a GitHub issue or create a pull request. 10 | 11 | ## Glossary Committers 12 | 13 | All issues and pull requests are evaluted by the Glossary Committers. The current committers are, in alphabetical order: 14 | 15 | * Alex Fazio, AI Garden (https://x.com/alxfazio) 16 | 17 | From time to time, the project will consider changing or expanding its committers. 18 | 19 | ## Working Group Chair 20 | 21 | The Chair of The A to Z of AI Working Group is: 22 | 23 | * Alex Fazio, AI Garden (https://x.com/alxfazio) 24 | 25 | ## Guidelines 26 | 27 | - To propose specific changes to a glossary entry, edit that entry in your branch and issue a pull request. 28 | - To request that an entry be clarified, updated or reconsidered, you may alternatively open an issue. 29 | - To propose a new entry to the glossary, either create an issue or, if you have the definition drafted, add the entry to your branch (in its proper alphabetical location) and create a pull request. 30 | 31 | ## Issues & Pull Requests 32 | 33 | - Please jump in and help us out by reviewing open issues and making pull requests to resolve them. 34 | - If you see anything you think can be improved, feel free to open an issue or, even, better submit your suggestions via a pull request. 35 | - Please title your pull requests appropriately, summing up what your commits are about. 36 | - For new terms, make sure that your entry doesn't already exist somewhere else in the glossary. 37 | 38 | Requests: 39 | 40 | - For glossary entry changes and additions, please raise separate pull requests for each entry. This makes it easier to discuss and shepherd updates in self-contained units. 41 | - Please provide links to third party uses that support your issue or pull request. 42 | 43 | ## Philosophy 44 | 45 | - The A to Z of AI seeks to be concise, including words that are unique to Artificial Intelligence, or which have different meanings when used in an AI context. 46 | 47 | ## Descriptive vs Prescriptive 48 | 49 | - In most circumstances, the definition of a term should be based on empirical evidence of contemporary usage as published in literature, academic articles, talks and white papers. 50 | - In some cases, a term will suffer from conflation, imprecise usage, or, even worse, outright conflicting defintions. In these cases, the Glossary Committers will consider proposed clarifications or focused definitions on a case-by-case basis. 51 | -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # The A to Z of AI 2 | ### AI terms, from “Alignment” to “Zero-Shot Prompting”, explained in plain English 3 | 4 |
5 |
6 |