├── README.md ├── archived-201510-201806.md ├── conferences_recap └── emnlp2018_recap.md └── docs ├── _config.yml └── index.html /README.md: -------------------------------------------------------------------------------- 1 | # NLP Reading Group 2 | 3 | The target audience is all the members of the NLP group and other possible interested participants. 4 | 5 | The meeting takes place weekly during term for one hour usually on **Mondays, 14:00-15:00 UK Time**. See below a list of the upcoming and past meetings. 6 | 7 | The meetings of the group are informal and no necessary preparation will be required with the exception of the moderator reading the current paper and the rest having at least a brief overview of it. 8 | 9 | There is also a #readinggroup channel in the NLP group's unofficial Slack channel: https://usfd-nlp.slack.com/messages 10 | 11 | A list of past meetings before 2018/19 can be found [here](https://www.sheffield.ac.uk/dcs/research/groups/nlp#tab05). 12 | 13 | 16 | 17 | ----- 18 | 19 | 23 | 24 | **Upcoming Meeting(s)** 25 | --------------- 26 | 27 | Will restart after summer break. 28 | 29 | Past Meetings 30 | --------------- 31 | * **Mon 30 May 2022** 32 | - **Paper:** [Context-aware Decoder for Neural Machine Translation using a Target-side Document-Level Language Model)](https://arxiv.org/abs/2010.12827) 33 | - **Moderator:** Sebastian Vincent 34 | 35 | * **Mon 16 May 2022** 36 | - **Paper:** [Word Order Does Matter (And Shuffled Language Models Know It)](https://arxiv.org/pdf/2203.10995.pdf) 37 | - **Moderator:** Edward Gow-Smith 38 | 39 | * **Mon 9 May 2022** 40 | - **Paper:** [Chain of Thought Prompting Elicits Reasoning in Large Language Models](https://arxiv.org/abs/2201.11903) 41 | - **Moderator:** Jasivan A Sivakumar 42 | 43 | * **Mon 11th Apr 2022** 44 | - **Paper:** [Refocusing on Relevance: Personalization in NLG]([https://aclanthology.org/2021.bionlp-1.5/](https://arxiv.org/pdf/2109.05140.pdf)) 45 | - **Moderator:** Tomas Goldsack 46 | 47 | * **Mon 4th Apr 2022** 48 | - **Paper:** [Are we there yet? Exploring clinical domain knowledge of BERT models](https://aclanthology.org/2021.bionlp-1.5/) 49 | - **Moderator:** Dylan RS Phelps 50 | 51 | * **Mon 7th Mar 2022** 52 | - **Paper:** [I Wish I Would Have Loved This One, But I Didn't -- A Multilingual Dataset for Counterfactual Detection in Product Reviews](https://arxiv.org/pdf/2101.00403.pdf) 53 | - **Moderator:** Mali Jin 54 | 55 | * **Mon 7th Mar 2022** 56 | - **Paper:** [Superbizarre Is Not Superb: Derivational Morphology Improves BERT's Interpretation of Complex Words](https://arxiv.org/pdf/2101.00403.pdf) 57 | - **Moderator:** Edward Gow-Smith 58 | 59 | * **Mon 28th Feb 2022** 60 | - **Paper:** [Text Classification Using Label Names Only: A Language Model Self-Training Approach](https://aclanthology.org/2020.emnlp-main.724.pdf) 61 | - **Moderator:** Danae Sanchez Villegas 62 | 63 | * **Mon 21st Feb 2022** 64 | - **Paper:** [Finetuned Language Models Are Zero-Shot Learners](https://openreview.net/forum?id=gEZrGCozdqR) 65 | - **Moderator:** Katerina Margatina 66 | 67 | * **Mon 14th Feb 2022** 68 | - **Paper:** [ProFormer: Towards On-Device LSH Projection Based Transformers](https://aclanthology.org/2021.eacl-main.246.pdf) 69 | - **Moderator:** Huiyin 70 | 71 | * **Mon 07th Feb 2022** 72 | - **Paper:** [Structurizing Misinformation Stories via Rationalizing Fact-Checks](https://aclanthology.org/2021.acl-long.51/) 73 | - **Moderator:** Cass Zhao 74 | 75 | * **Mon 31st Jan 2022** 76 | - **Paper:** [Data2vec: A General Framework for Self-supervised Learning in Speech, Vision and Language](https://ai.facebook.com/research/data2vec-a-general-framework-for-self-supervised-learning-in-speech-vision-and-language) 77 | - **Moderator:** Xutan 78 | 79 | * **Mon 24th Jan 2022** 80 | - **Paper:** [A Survey on Dialogue Summarization: Recent Advances and New Frontiers](https://arxiv.org/abs/2107.03175) 81 | - **Moderator:** Guanyu Huang 82 | 83 | * **Mon 06th Dec 2021** 84 | - **Paper:** 85 | - **Moderator:** Zeerak 86 | - 87 | * **Mon 29th Nov 2021** 88 | - **Paper:** 89 | - **Moderator:** Dylan 90 | 91 | * **Mon 22nd Nov 2021** ![Classic](https://img.shields.io/badge/Classic-Old%20NLP--relevant%20papers-red) 92 | - **Paper:** [ Structure Mapping in Analogy and Similarity](https://groups.psych.northwestern.edu/gentner/papers/GentnerMarkman97.pdf) *(American Psychologist 1997)* 93 | - **Moderator:** Xutan 94 | 95 | * **Mon 08th Nov 2021** 96 | - Review Session for ACL Submissions 97 | 98 | * **Mon 15th Nov 2021** 99 | - ACL Deadline 100 | 101 | * **Mon 18th Oct 2021** 102 | - **Paper:** [The R-U-A-Robot Dataset: Helping Avoid Chatbot Deception by Detecting User Questions About Human or Non-Human Identity](https://aclanthology.org/2021.acl-long.544/) *(ACL'21)* 103 | - **Moderator:** Yida 104 | 105 | * **Mon 11th Oct 2021** 106 | - **Paper:** [Debugging Tests for Model Explanations](https://papers.nips.cc/paper/2020/file/075b051ec3d22dac7b33f788da631fd4-Paper.pdf) 107 | - **Moderator:** George 108 | 109 | * **Mon 04th Oct 2021** 110 | - **Paper:** [Predicting emergent linguistic compositions through time: Syntactic frame extension via multimodal chaining 111 | ](https://arxiv.org/abs/2109.04652) *(EMNLP'21)* 112 | - **Moderator:** Danae 113 | 114 | * **Mon 27th Sep 2021** 115 | - **Paper:** [Does Pretraining for Summarization Require Knowledge Transfer?](https://arxiv.org/abs/2109.04953) *(EMNLP'21 - Findings)* 116 | - **Moderator:** Katerina 117 | 118 | * **Mon 20th Sep 2021** 119 | - **Paper:** [Paragraph-level Simplification of Medical Texts](https://aclanthology.org/2021.naacl-main.395/) *(NAACL'21)* 120 | - **Moderator:** Fernando 121 | 122 | * **Mon 16th Aug 2021** 123 | - **Paper:** [From characters to words: the turning point of BPE merges](https://aclanthology.org/2021.eacl-main.302.pdf) *(EACL'21)* 124 | - **Moderator:** Xutan 125 | 126 | * **Mon 28th June 2021** 127 | - **Paper:** [Can a Fruit Fly Learn Word Embeddings?](https://openreview.net/pdf?id=xfmSoxdxFCG) *(ICLR'21)* 128 | - **Moderator:** Yida M 129 | 130 | * **Mon 21st June 2021** 131 | - **Paper:** [Does syntax matter? A strong baseline for Aspect-based Sentiment Analysis with RoBERTa](https://www.aclweb.org/anthology/2021.naacl-main.146/) *(NAACL'21)* 132 | - **Moderator:** Mali 133 | 134 | * **Mon 14th June 2021** 135 | - **Paper:** [Learning The Difference That Makes A Difference With Counterfactually-Augmented Data](https://openreview.net/forum?id=Sklgs0NFvr) *(ICLR'20)* 136 | - **Moderator:** Katerina 137 | 138 | * **Mon 7th June 2021** 139 | - **Paper:** - 140 | - **Moderator:** - 141 | 142 | * **Mon 26th April 2021** 143 | - **Paper:** [Estimating predictive uncertainty for rumour verification models](https://www.aclweb.org/anthology/2020.acl-main.623.pdf) *(ACL'20)* 144 | - **Moderator:** Wenzhe 145 | 146 | * **Mon 22nd March 2021** 147 | - **Paper:** [Summarising Historical Text in Modern Languages](https://arxiv.org/abs/2101.10759) *(EACL'21)* 148 | - **Moderator:** Xutan 149 | 150 | * **Mon 15th March 2021** 151 | - **Paper:** [LAMBDA NETWORKS: MODELING LONG-RANGE INTERACTIONS WITHOUT ATTENTION](https://openreview.net/pdf?id=xTJEN-ggl1b) 152 | - **Moderator:** George 153 | 154 | * **Mon 8th March 2021** 155 | - **Paper:** [Grounded Conversation Generation as Guided Traverses in Commonsense Knowledge Graphs](https://www.aclweb.org/anthology/2020.acl-main.184.pdf) 156 | - **Moderator:** Ruizhe Li 157 | 158 | 159 | * **Mon 1st March 2021** 160 | - **Paper:** [CogLTX: Applying BERT to Long Texts](https://keg.cs.tsinghua.edu.cn/jietang/publications/NIPS20-Ding-et-al-CogLTX.pdf) *(NeurIPS'20)* 161 | - **Moderator:** Yida Mu 162 | 163 | * **Mon 22nd Feb 2021** 164 | - **Paper:** [Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting](https://arxiv.org/pdf/2012.07436.pdf) *(AAAI'21 Best Paper)* 165 | - **Moderator:** Xutan 166 | 167 | * **Mon 8th Feb 2021** 168 | - **Paper:** [Cross-Media Keyphrase Prediction: A Unified Framework with Multi-Modality Multi-Head Attention and Image Wordings](https://www.aclweb.org/anthology/2020.emnlp-main.268/) 169 | - **Moderator:** Danae 170 | 171 | * **Fri 29th Jan 2021** 172 | 173 | - **ACL Review/Feedback session** 174 | * **Mon 18th Jan 2021** 175 | - **Paper:** [Affective and Contextual Embedding for Sarcasm Detection](https://www.aclweb.org/anthology/2020.coling-main.20/) 176 | - **Moderator:** Mali 177 | 178 | * **Mon 11th Jan 2021** 179 | - **Paper:** [Unsupervised Data Augmentation for Consistency Training](https://openreview.net/forum?id=ByeL1R4FvS) 180 | - **Moderator:** Katerina 181 | 182 | * **Mon 4th Jan 2021** 183 | - **Paper:** [A Diagnostic Study of Explainability Techniques for Text Classification](https://www.aclweb.org/anthology/2020.emnlp-main.263/) 184 | - **Moderator:** George 185 | 186 | * **Mon 16th Nov 2020** 187 | - **Paper:** [Heterogeneous Graph Neural Networks for Extractive Document Summarization](https://arxiv.org/pdf/2004.12393.pdf)[Slide](https://docs.google.com/presentation/d/1SlaqNVTvOorKmp3a6Zsd7OpDLiUJVFtCG48szc-rOV0/edit?usp=sharingv) 188 | - **Moderator:** Wenzhe 189 | 190 | * **Mon 9th Nov 2020** 191 | - **Paper:** [Self-Attention Guided Copy Mechanism for Abstractive Summarization](https://www.aclweb.org/anthology/2020.acl-main.125.pdf) 192 | - **Moderator:** Hardy 193 | 194 | * **Mon 2th Nov 2020** 195 | - **Paper:** [Where Are the Facts? Searching for Fact-checked Information to Alleviate the Spread of Fake News](https://arxiv.org/pdf/2010.03159.pdf) 196 | - **Moderator:** Yida Mu 197 | 198 | * **Mon 26th Oct 2020** 199 | - **Paper:** [Cold-start Active Learning through Self-supervised Language Modeling](https://arxiv.org/abs/2010.09535) 200 | - **Moderator:** Katerina 201 | 202 | * **Mon 19th Oct 2020** 203 | - **Paper:** [What Was Written vs. Who Read It: News Media Profiling Using Text Analysis and Social Media Context](https://www.aclweb.org/anthology/2020.acl-main.308.pdf) 204 | - **Moderator:** Danae 205 | 206 | * **Mon 12th Oct 2020** 207 | - **Paper:** [Joint Modelling of Emotion and Abusive Language Detection](https://arxiv.org/pdf/2005.14028.pdf) 208 | - **Moderator:** Mali 209 | 210 | * **Mon 5th Oct 2020** 211 | - EACL Paper Review Session 212 | 213 | * **Mon 28th Sep 2020** 214 | - **Paper:** [Towards Faithfully Interpretable NLP Systems](https://arxiv.org/pdf/2004.03685.pdf) 215 | - **Moderator:** George 216 | 217 | * **Mon 20th July 2020** 218 | 219 | - **Paper:** [Longformer: The Long-Document Transformer](https://arxiv.org/abs/2004.05150) 220 | - **Moderator:** Xutan 221 | 222 | * **Mon 13th July 2020** 223 | 224 | - **Paper:** [Latent Space Factorisation and Manipulation via Matrix Subspace Projection](https://xiao.ac/_data/msp/MSP-icml2020-near-camera-ready.pdf) 225 | - **Moderator:** Ruizhe 226 | 227 | * **Mon 29 June 2020** 228 | 229 | - **Paper:** [Context-aware monolingual repair for neural machine translation](https://www.mendeley.com/catalogue/264b2ffe-6877-3d9b-86d2-ede3bd4caf55/?utm_source=desktop&utm_medium=1.19.4&utm_campaign=open_catalog&userDocumentId=%7B5f02f55a-058b-41d7-95fd-576d293ed801%7D) 230 | - **Moderator:** Sebastian 231 | 232 | * **Mon 22 June 2020** 233 | 234 | - **Paper:** [Relational inductive biases, deep learning, and graph networks](https://arxiv.org/abs/1806.01261) 235 | - **Moderator:** Peter 236 | 237 | * **Mon 15 June 2020** 238 | 239 | - **Paper:** [Multimodal Quality Estimation for Machine Translation]() 240 | - **Moderator:** Shu Okabe / Fred 241 | 242 | * **Mon 8 June 2020** 243 | 244 | - **Paper:** [Analyzing Political Parody in Social Media](https://arxiv.org/abs/2004.13878) 245 | - **Moderator:** Danae 246 | 247 | * **Fri 29 May 2020** 248 | 249 | - **EMNLP Review/Feedback session** 250 | 251 | 252 | * **Mon 18 May 2020** 253 | 254 | - **Paper:** [Learning to Faithfully Rationalize by Construction](https://arxiv.org/abs/2005.00115) 255 | - **Moderator:** George 256 | 257 | 258 | * **Mon 11 May 2020** 259 | 260 | - **Paper:** [Aspect Sentiment Classification Towards Question-Answering with Reinforced Bidirectional Attention Network](https://www.aclweb.org/anthology/P19-1345.pdf). 261 | - **Moderator:** Mali 262 | - **Room:** Google Hangouts 263 | 264 | * **Mon 04 May 2020** 265 | 266 | - **Paper:** [BLEU might be Guilty but References are not Innocent](https://arxiv.org/pdf/2004.06063.pdf) 267 | - **Moderator:** Fernando 268 | - **Room:** Google Hangouts 269 | 270 | * **Mon 27 Apr 2020** 271 | 272 | - **Paper:** [“Trust me, I have a Ph.D.”: A Propensity Score Analysis on the Halo Effect ofDisclosing One’s Offline Social Status in Online Communities](https://arxiv.org/pdf/2004.00105.pdf) (ICWSM-20) 273 | - **Moderator:** Yida 274 | - **Room:** Google Hangouts 275 | 276 | * **Mon 20 Apr 2020** 277 | 278 | - **Paper:** [How to (Properly) Evaluate Cross-Lingual Word Embeddings: On Strong Baselines, Comparative Analyses, and Some Misconceptions](https://www.aclweb.org/anthology/P19-1070/) (ACL'19) 279 | - **Moderator:** Xutan 280 | - **Room:** Google Hangouts 281 | 282 | * **Mon 6 Apr 2020** 283 | 284 | - **Paper:** [Many Faces of Feature Importance: Comparing Built-in and Post-hocFeature Importance in Text Classification](https://www.aclweb.org/anthology/D19-1046.pdf) 285 | - **Moderator:** George 286 | 287 | * ~~**Mon 16 Mar 2020**~~ **Now Mon 30 Mar 2020 (due to COVID-19)** 288 | 289 | - **Paper:** [Lagging Inference Networks and Posterior Collapse in Variational Autoencoders](https://arxiv.org/pdf/1901.05534.pdf) 290 | - **Moderator:** Ruizhe 291 | - **Room:** G12-Blue 292 | 293 | * **Mon 9 Mar 2020** 294 | 295 | - **Paper:** [BatchBALD: Efficient and Diverse Batch Acquisition for Deep Bayesian Active Learning]( https://papers.nips.cc/paper/8925-batchbald-efficient-and-diverse-batch-acquisition-for-deep-bayesian-active-learning.pdf) 296 | - **Moderator:** Katerina 297 | - **Room:** G12-Blue 298 | 299 | * **Mon 2 Mar 2020** 300 | 301 | - **Paper:** [Integrating Text and Image: Determining Multimodal Document Intent in Instagram Posts](https://www.aclweb.org/anthology/D19-1469.pdf) 302 | - **Moderator:** Danae 303 | - **Room:** G12-Blue 304 | 305 | * **Mon 24 Feb 2020** 306 | 307 | - **Paper:** [Interpretable emoji prediction via label-wise attention LSTMs](https://www.aclweb.org/anthology/D18-1508.pdf) 308 | - **Moderator:** George 309 | - **Room:** G12-Blue 310 | 311 | 312 | * **Mon 17 Feb 2020** 313 | - **Paper:** [Detecting Customer Complaint Escalation with Recurrent Neural Networks and Manually-Engineered Features](https://www.aclweb.org/anthology/N19-2008.pdf) 314 | - **Moderator:** Mali 315 | - **Room:** G12-Blue 316 | 317 | 318 | * **Mon 10 Feb 2020** 319 | 320 | - **Paper:** Hubless Nearest Neighbor Search for Bilingual Lexicon Induction (ACL 19) ([paper](https://www.aclweb.org/anthology/P19-1399/) | [code](https://github.com/baidu-research/HNN)) 321 | - **Moderator:** Xutan 322 | - **Room:** G12-Blue 323 | 324 | * **Mon 3 Feb 2020** 325 | 326 | - **Paper:** [Affect-Driven Dialog Generation](https://www.aclweb.org/anthology/N19-1374.pdf) 327 | - **Moderator:** Danae 328 | - **Room:** G12-Blue 329 | 330 | * **Mon 27 Jan 2020** 331 | 332 | - **Paper:** [BERTScore: Evaluating Text Generation with BERT](https://arxiv.org/abs/1904.09675) 333 | - **Moderator:** Fernando 334 | - **Room:** G12-Blue 335 | 336 | 337 | * **Mon 20 Jan 2020** 338 | - **Paper:** [Uncertainty-Aware Attention for Reliable Interpretation and Prediction](https://papers.nips.cc/paper/7370-uncertainty-aware-attention-for-reliable-interpretation-and-prediction.pdf) 339 | - **Moderator:** George 340 | - **Room:** G12-Blue 341 | 342 | 343 | 344 | * **Fri 06 Dec 2019** 345 | 346 | - ***ACL Review/Feedback session* 347 | - **Room:** G12-Blue 348 | 349 | * **Mon 02 Dec 2019** 350 | 351 | - **Paper:** 352 | - **Moderator:** Cancel due to ACL deadline 353 | - **Room:** G12-Blue 354 | 355 | 356 | 357 | 358 | * **Mon 25 Nov 2019** 359 | 360 | - **Paper:** [What Does This Word Mean? Explaining Contextualized Embeddings with Natural Language Definition](https://www.aclweb.org/anthology/D19-1627.pdf) 361 | - **Moderator:** Varvara 362 | - **Room:** ~~G12-Blue~~ COM-108 - Ada Lovelace Room 363 | 364 | 365 | 366 | 367 | * **Mon 18 Nov 2019** 368 | 369 | - **Paper:** [Hierarchical Transfer Learning for Multi-label Text Classification](https://www.aclweb.org/anthology/P19-1633.pdf) 370 | - **Moderator:** Mali 371 | - **Room:** G12-Blue 372 | 373 | * **Mon 11 Nov 2019** 374 | 375 | - **Paper:** Incorporating Priors with Feature Attribution on Text Classification (https://www.aclweb.org/anthology/P19-1631) 376 | - **Moderator:** - 377 | - **Room:** G12-Blue 378 | 379 | * **Mon 4 Nov 2019** 380 | 381 | - **Paper:** DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter. NeurIPS workshop paper 2019 ([paper](https://arxiv.org/pdf/1910.01108.pdf)|[blog](https://medium.com/huggingface/distilbert-8cf3380435b5)|[slides](https://drive.google.com/file/d/1b6dfEgmXJteuFLa2DuWivzSpOiplQq4P/view)) 382 | - **Moderator:** Katerina 383 | - **Room:** G12-Blue 384 | 385 | * **Mon 28 Oct 2019** 386 | 387 | - **Paper:** A robust self-learning method for fully unsupervised cross-lingual mappings of word embeddings 388 | ([link](https://www.aclweb.org/anthology/P18-1073.pdf)|[video](https://vimeo.com/285800964)|[slides](https://www.aclweb.org/anthology/attachments/P18-1073.Presentation.pdf)) 389 | - **Moderator:** Xutan 390 | - **Room:** G12-Blue 391 | 392 | * **Mon 21 Oct 2019** 393 | 394 | - **Paper:** Fine-Grained Analysis of Propaganda in News Articles (https://arxiv.org/pdf/1910.02517.pdf) 395 | - **Moderator:** Yida 396 | - **Room:** G12-Blue 397 | 398 | * **Mon 14 Oct 2019** 399 | 400 | - **Paper:** [MeanSum: A Neural Model for Unsupervised Multi-document Abstractive Summarization](http://proceedings.mlr.press/v97/chu19b/chu19b.pdf) 401 | - **Moderator:** Hardy 402 | - **Room:** G12-Blue 403 | 404 | 405 | * **Mon 7 Oct 2019** 406 | 407 | - **Paper:** [We need to talk about standard splits](https://www.aclweb.org/anthology/P19-1267/) (ACL 2019 - Outstanding Paper Award) 408 | - **[Approximate Statistical Tests for Comparing Supervised Classification Learning Algorithms](https://www.mitpressjournals.org/doi/pdf/10.1162/089976698300017197)** 409 | - **Moderator:** Fernando 410 | - **Room:** G12-Blue 411 | 412 | 413 | * **Mon 30 Sep 2019** 414 | 415 | - **Paper:** Weight Uncertainty in Neural Networks (Blundell et al. 2015) (http://proceedings.mlr.press/v37/blundell15.pdf) 416 | - **Moderator:** George 417 | - **Room:** G12-Blue 418 | 419 | 420 | * **Mon 8 Jul 2019** 421 | 422 | - **Paper:** Tay et al. KDD 2018. Multi-Pointer Co-Attention Networks for Recommendation. (https://arxiv.org/abs/1801.09251) 423 | - **Moderator:** Haiyang 424 | - **Room:** G12-Blue 425 | 426 | 427 | * **Mon 1 Jul 2019&& 428 | 429 | - Early Rumour Detection (Zhou et al., 2019) NAACL 2019. (https://www.aclweb.org/anthology/N19-1163) 430 | - **Moderator:** Yida 431 | - **Room:** G12-Blue 432 | 433 | 434 | 435 | 436 | * **Mon 24 Jun 2019** 437 | 438 | - **Paper:** (Mao et al., 2018), ACL -- [Word Embedding and WordNet Based Metaphor Identification and Interpretation](https://www.aclweb.org/anthology/P18-1113) 439 | - **Moderator:** Varvara 440 | - **Room:** G12-Blue 441 | 442 | 443 | 444 | 445 | * **Mon 17 Jun 2019&& 446 | 447 | - **Paper:** What’s in a Name? Reducing Bias in Bios Without Access to Protected Attributes 448 | Alexey Romanov, Maria De-Arteaga, Hanna Wallach, Jennifer Chayes, Christian Borgs, Alexandra Chouldechova, Sahin Geyik, Krishnaram Kenthapadi, Anna Rumshisky and Adam Kalai (https://arxiv.org/abs/1904.05233v1) 449 | - **Moderator:** Zeerak 450 | - **Room:** TBA 451 | 452 | * **Mon 10 Jun 2019** 453 | 454 | - TBD 455 | - **Moderator:** George 456 | - **Room:** G12-Blue 457 | 458 | 459 | 460 | * **Mon 3 Jun 2019** 461 | 462 | - (\*ACL) Paper writing tips 463 | - **Moderator:** Nikos 464 | - **Room:** G12-Blue 465 | 466 | 467 | 468 | 469 | 470 | * **Fri 17 May 2019** 471 | 472 | - EMNLP Review session 473 | 474 | 475 | * **Mon 13 May 2019** 476 | 477 | - **Paper:** (Lei et al., 2017), ArXiv -- [Simple Recurrent Units for Highly Parallelizable Recurrence](https://arxiv.org/abs/1709.02755) 478 | - **Moderator:** Abiola 479 | - **Room:** G25 480 | 481 | * ~~**Mon 06 May 2019**~~ 482 | **Cancelled: Early May bank holiday** 483 | 484 | - **Paper:** TBA 485 | - **Moderator:** TBA 486 | - **Room:** TBA 487 | 488 | * **Mon 29 Apr 2019** 489 | 490 | - **Paper:** TBA 491 | - **Moderator:** Albiola 492 | - **Room:** TBA 493 | 494 | * ~~**Mon 22 Apr 2019**~~ 495 | **Cancelled: Easter Monday** 496 | 497 | - **Paper:** TBA 498 | - **Moderator:** TBA 499 | - **Room:** TBA 500 | 501 | * **Mon 15 Apr 2019** 502 | 503 | - **Paper:** Tolga Bolukbasi, Kai-Wei Chang, James Zou, Venkatesh Saligrama, and Adam Kalai. 2016. Man is to computer programmer as woman is to homemaker? Debiasing word embeddings. (https://dl.acm.org/citation.cfm?id=3157584) 504 | - **Moderator:** Alison 505 | - **Room:** TBA 506 | 507 | * **Mon 08 Apr 2019** 508 | 509 | - **Paper:** 510 | Identifying and understanding user reactions to deceptive and trusted social news sources 511 | (https://arxiv.org/pdf/1805.12032.pdf) 512 | How Humans versus Bots React to Deceptive and Trusted News Sources: A Case Study of Active Users 513 | (https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8508700) 514 | - **Moderator:** Yida 515 | - **Room:** G12-Blue 516 | 517 | * **Mon 01 Apr 2019** 518 | 519 | - **Paper:** Sarthak Jain and Byron C. Wallace (ArXiv) [Attention is not Explanation] (https://arxiv.org/pdf/1902.10186.pdf) 520 | - **Moderator:** George 521 | - **Room:** G12-Blue (Lewing Lab) 522 | 523 | 524 | 525 | * **Mon 25 Mar 2019** 526 | 527 | - **Paper:** 528 | Rebekah Overdorf, Bogdan Kulynych, Ero Balsa, Carmela Troncoso, Seda Gürses (ArXiv) [POTs: Protective Optimization Technologies](https://arxiv.org/abs/1806.02711) 529 | - **Moderator:** Zeerak 530 | - **Room:** G12-Blue (Lewin Lab) 531 | 532 | * **Mon 18 Mar 2019** 533 | 534 | - **Paper:** [Coherence Aware Topic Modeling (EMNLP 2018) ](http://aclweb.org/anthology/D18-1096) 535 | - **Moderator:** Areej 536 | - **Room:** G22-Orange 537 | 538 | * **Mon 11 Mar 2019** 539 | 540 | - **Paper:** Gregor Wiedemann, Eugen Ruppert, Raghav Jindal, Chris Biemann (GermEval, 2018) [Transfer Learning from LDA to BiLSTM-CNN for Offensive Language Detection in Twitter](https://arxiv.org/abs/1811.02906) 541 | - **Moderator:** Cass 542 | - **Room:** G12-Blue (Lewin Lab) 543 | 544 | * **Fri 01 Mar 2019** 545 | 546 | - ACL Review Session 547 | - **Time:** 13:00-15:00 548 | - **Room:** G12-Blue (Lewin Lab) 549 | 550 | 551 | 552 | * **Mon 25 Feb 2019** 553 | 554 | - **Paper:** (Leila Arras et al, 2017), ACL -- [Explaining Recurrent Neural Network Predictions in Sentiment Analysis](http://www.aclweb.org/anthology/W17-5221) 555 | - **Moderator:** George 556 | - **Room:** G25 557 | 558 | 559 | * **Mon 18 Feb 2019** 560 | 561 | - **Paper:** (Dehghani et al., 2019), ArXiv -- [Universal Transformers](https://arxiv.org/abs/1807.03819) 562 | - **Moderator:** Hardy 563 | - **Room:** G25 564 | 565 | * **Mon 11 Feb 2019** 566 | 567 | - **Paper:** (Houlsby et al., 2019), ArXiv -- [Parameter-Efficient Transfer Learning for NLP](https://arxiv.org/abs/1902.00751) 568 | - **Moderator:** Fred 569 | - **Room:** G12-Blue 570 | 571 | * **Mon 04 Feb 2019** 572 | 573 | - **Paper:** Howard and Ruder (ACL, 2018) [Universal Language Model Fine-tuning for Text Classification](http://aclweb.org/anthology/P18-1031) 574 | - **Moderator:** Nikos 575 | - **Room:** G25 576 | 577 | * **Mon 28 Jan 2019** 578 | 579 | - **Paper:** Surya et al. (2019). [Unsupervised Neural Text Simplification](https://arxiv.org/abs/1810.07931) 580 | - **Moderator:** Fernando 581 | - **Room:** G25 582 | 583 | * **Wed 12 Dec 2018** 584 | 585 | - **Paper:** 586 | * Wu etal (2018): Word Mover's Embedding: From Word2Vec to Document Embedding https://arxiv.org/abs/1811.01713 587 | * Wu etal (2018): D2KE: From Distance to Kernel and Embedding 588 | - **Moderator:** Johann 589 | - **Room:** COM-G25 590 | 591 | 592 | * **Wed 05 Dec 2018** 593 | 594 | - NAACL submissions review (deadline: Dec 10) 595 | - **Room:** G-25 596 | 597 | 598 | * **Wed 28 Nov 2018** 599 | * **Paper:** I'd like to try and look at the following three because they are all somewhat related, but with the focus on BERT and the transformer-based architectures and representations: 600 | * Devlin etal (2018): BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding https://arxiv.org/abs/1810.04805 601 | * Cer etal (2018): Universal Sentence Encoder https://arxiv.org/abs/1803.11175 602 | * Yang etal (2018): Learning Semantic Textual Similarity from Conversations https://arxiv.org/abs/1804.07754 603 | * Note: it will help to know the paper: Vaswani etal (2017): Attention is all you need https://arxiv.org/abs/1706.03762 604 | * **Moderator:** Johann 605 | * **Room:** COM-G25 606 | 607 | 608 | 609 | * **Wed 21 Nov 2018** 610 | * **Paper:** Peters et al (2018), [Deep contextualized word representations](http://aclweb.org/anthology/N18-1202), NAACL 2018 611 | * **Moderator:** Carol 612 | * **Room:** COM-G25 613 | * Related papers mentioned: 614 | * word senses in embeddings: Arora etal 2016: Linear Algebraic Structure of Word Senses, with Applications to Polysemy. http://arxiv.org/abs/1601.03764 615 | * combining different vector spaces: Coates and Bollegala 2018: Frustratingly Easy Meta-Embedding -- Computing Meta-Embeddings by Averaging Source Word Embeddings http://aclweb.org/anthology/N18-2031 616 | * BPEs for MT: Sennrich et al. (2016): Neural Machine Translation of Rare Words with Subword Units. http://www.aclweb.org/anthology/P16-1162 617 | 618 | * **Wed 14 Nov 2018** 619 | 620 | - **Paper:** EMNLP recap: [papers presented](conferences_recap/emnlp2018_recap.md) 621 | - **Moderator:** Zeerak 622 | - **Room:** COM-G22 Blue 623 | 624 | 625 | * **Wed 07 Nov 2018** 626 | 627 | - **NLP seminar (RG is postponed)** 628 | 629 | 630 | 631 | * **Wed 31 Oct 2018** 632 | 633 | - **Paper:** Zheng et al (2018), [Multi-Reference Training with Pseudo-References for Neural Translation and Text Generation](https://arxiv.org/pdf/1808.09564.pdf), in EMNLP 2018 634 | - **Moderator:** Makis 635 | - **Room:** COM-G25 636 | 637 | 638 | 639 | 640 | * **Wed 24 Oct 2018** 641 | - **Paper:** Xing and Paul (2018), [Diagnosing and Improving Topic Models by Analyzing Posterior Variability](https://www.aaai.org/ocs/index.php/AAAI/AAAI18/paper/view/16213/16168), In AAAI 642 | - **Moderator:** Areej 643 | - **Room:** COM-G25 644 | 645 | 646 | 647 | 648 | * **Wed 17 Oct 2018** 649 | 650 | - **Paper:** Dong, Quirk and Lapata (2018), [Confidence Modeling for Neural Semantic Parsing](https://arxiv.org/pdf/1805.04604.pdf), In ACL 651 | - **Moderator:** Nikos 652 | - **Room:** COM-G25 653 | 654 | 655 | 656 | * **Wed 10 Oct 2018** 657 | 658 | - Kick-off meeting 659 | - **Room:** COM-G22 Blue 660 | -------------------------------------------------------------------------------- /archived-201510-201806.md: -------------------------------------------------------------------------------- 1 | # READING GROUP: October 2015 -- June 2018 2 | 3 | (converted copy of the page https://www.sheffield.ac.uk/dcs/research/groups/nlp#tab05) 4 | 5 | The target audience is all the members of the NLP group and other possible interested participants. 6 | 7 | The meeting will take place **weekly for one hour** usually on **Tuesdays from 11-12pm**. 8 | 9 | The meetings of the group will be informal and no necessary preparation will be required with the exception of the moderator reading the current paper and the rest having at least a brief overview of it. 10 | 11 | ## Next Meeting 12 | 13 | ## Past Meetings 14 | 15 | **Tuesday 12 June 2018** 16 | 17 | [Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks](https://arxiv.org/pdf/1703.03400.pdf) 18 | 19 | Chelsea Finn, Pieter Abbeel, Sergey Levine, ICML 2017 20 | [Blog post about the paper by the authors](http://bair.berkeley.edu/blog/2017/07/18/learning-to-learn/) 21 | 22 | **Tuesday 10 April 2018** 23 | 24 | [Style Transfer from Non-Parallel Text by Cross-Alignment](https://papers.nips.cc/paper/7259-style-transfer-from-non-parallel-text-by-cross-alignment.pdf) 25 | 26 | Shen, T; Lei, T; Barzilay, R; Jaakola, T. 27 | 28 | **Tuesday 3 April 2018** 29 | 30 | [Generating Natural Adversarial Examples](https://arxiv.org/pdf/1710.11342.pdf) 31 | 32 | Zhengli Zhao, Dheeru Dua and Sameer Singh 33 | 34 | **Tuesday 20 February 2018** 35 | 36 | ACL Paper submission feedback session 37 | 38 | **Tuesday 13 February 2018** 39 | 40 | [Unbounded cache model for online language modeling with open vocabulary](https://arxiv.org/pdf/1711.02604.pdf) 41 | 42 | Edouard Grave, Moustapha Cisse & Armand Joulin 43 | 44 | **Tuesday 6 February 2018** 45 | 46 | [Neural Sequence Learning Models for Word Sense Disambiguation](http://wwwusers.di.uniroma1.it/~raganato/pubs/emnlp17_raganatoetal.pdf) 47 | 48 | Alessandro Raganato, Claudio Delli Bovi & Roberto Navigli 49 | 50 | **Tuesday 30 January 2018** 51 | 52 | [End-to-End Differentiable Proving](https://arxiv.org/abs/1705.11040) 53 | 54 | Tim Rocktäschel & Sebastian Riedel 55 | 56 | **Tuesday 23 January 2018** 57 | 58 | Unsupervised Learning of Universal Sentence Representations from NLI Data. 59 | 60 | * [paper](https://arxiv.org/abs/1705.02364) 61 | * [code](https://github.com/facebookresearch/InferSent) 62 | 63 | **Tuesday 28 November 2017** 64 | 65 | [Google’s Multilingual Neural Machine Translation System: Enabling Zero-Shot Translation](http://aclweb.org/anthology/Q/Q17/Q17-1024.pdf) 66 | 67 | Melvin Johnson, Mike Schuster, Quoc V. Le, et al. 68 | 69 | **Tuesday 14 November 2017** 70 | 71 | [Representations of language in a model of visually grounded speech signal](https://arxiv.org/pdf/1702.01991.pdf) 72 | 73 | Grzegorz Chrupała, Lieke Gelderloos & Afra Alishahi 74 | 75 | **Tuesday 7 November 2017** 76 | 77 | [A Class of Submodular Functions for Document Summarization](https://dl.acm.org/citation.cfm?id=2002537) 78 | 79 | Hui Lin & Jeff Bilmes 80 | 81 | **Tuesday 31 October 2017** 82 | 83 | [Question Generation for Question Answering](http://aclweb.org/anthology/D17-1091) 84 | 85 | Nan Duan, Duyu Tang, Peng Chen & Ming Zhou 86 | 87 | **Tuesday 24 October 2017** 88 | 89 | [Morphological Inflection Generation with Hard Monotonic Attention](http://www.aclweb.org/anthology/P17-1183) 90 | 91 | Roee Aharoni & Yoav Goldberg 92 | 93 | **Tuesday 17 October 2017** 94 | 95 | [A Factored Neural Network Model for Characterizing Online Discussions in Vector Space](http://www.aclweb.org/anthology/D17-1242) 96 | 97 | Hao Cheng, Hao Fang, Mari Ostendorf 98 | 99 | **Tuesday 10 October 2017** 100 | 101 | [Understanding Black-box Predictions via Influence Functions](http://proceedings.mlr.press/v70/koh17a.html) 102 | 103 | Pang Wei Koh, Percy Liang; Published in Proceedings of International Conference on Machine Learning, 2017 104 | 105 | **Tuesday 3 October 2017** 106 | 107 | [Zero-Shot Relation Extraction via Reading Comprehension](https://aclanthology.coli.uni-saarland.de/pdf/K/K17/K17-1034.pdf) 108 | 109 | Omer Levy, Minjoon Seo, Eunsol Choi and Luke Zettlemoyer 110 | 111 | **Tuesday 19 September 2017** 112 | 113 | "[Men also like shopping: Reducing Gender Bias Amplification Using Corpus Level Constraints](http://www.aclweb.org/anthology/D/D17/D17-1319.pdf)" 114 | 115 | **Tuesday 29 August 2017** 116 | 117 | [Deeply AggreVaTeD: Differentiable Imitation Learning for Sequential Prediction](http://proceedings.mlr.press/v70/sun17d.html) 118 | 119 | Wen Sun, Arun Venkatraman, Geoffrey J. Gordon, Byron Boots, J. Andrew Bagnell 120 | 121 | Proceedings of the 34th International Conference on Machine Learning, PMLR 70:3309-3318, 2017. 122 | 123 | **Tuesday 22 August 2017** 124 | 125 | [Split and Rephrase](https://arxiv.org/pdf/1707.06971.pdf), Accepted for EMNLP 2017 126 | 127 | Shashi Narayan, Claire Gardent, Shay B. Cohen and Anastasia Shimorina 128 | 129 | **Tuesday 15 August 2017** 130 | 131 | [Attention Is All You need](https://arxiv.org/pdf/1706.03762.pdf) 132 | A new simple network architecture, the Transformer, based solely on attention mechanisms, dispensing with recurrence and convolutions entirely 133 | 134 | **Tuesday 8 August 2017** 135 | 136 | [Learning to Compute Word Embeddings On the Fly](https://arxiv.org/pdf/1706.00286.pdf) 137 | 138 | Dzmitry Bahdanau, Tom Bosc, Stanisław Jastrzębski, Edward Grefenstette, Pascal Vincent, Yoshua Bengio 139 | 140 | **Tuesday 1 August 2017** 141 | 142 | [Learning to Generate Textual Data](https://aclweb.org/anthology/D16-1167), EMNLP 2016 143 | Guillaume Bouchard and Pontus Stenetorp and Sebastian Riedel 144 | 145 | **Tuesday 11 July 2017** 146 | 147 | [SoundNet: Learning Sound Representations from Unlabeled Video](https://projects.csail.mit.edu/soundnet/) 148 | 149 | Yusuf Aytar, Carl Vondrick, Antonio Torralba 150 | 151 | **Tuesday 4 July 2017** 152 | 153 | [Sentence Simplification with Deep Reinforcement Learning](https://arxiv.org/pdf/1703.10931.pdf) 154 | 155 | Xingxing Zhang, Mirella Lapata 156 | 157 | **Tuesday 27 June 2017** 158 | 159 | [Generation and Comprehension of Unambiguous Object Descriptions](https://arxiv.org/pdf/1511.02283) 160 | 161 | Junhua Mao, Jonathan Huang, Alexander Toshev, Oana Camburu, Alan Yuille, Kevin Murphy 162 | 163 | **Tuesday 20 June 2017** 164 | 165 | Understanding the BPE algorithm 166 | 167 | * [https://arxiv.org/pdf/1508.07909.pdf](https://arxiv.org/pdf/1508.07909.pdf) 168 | * [http://www.drdobbs.com/a-new-algorithm-for-data-compression/184402829?pgno=1](http://www.drdobbs.com/a-new-algorithm-for-data-compression/184402829?pgno=1) 169 | 170 | **Tuesday 13 June 2017** 171 | 172 | [Sequence-to-Sequence Models Can Directly Transcribe Foreign Speech](https://arxiv.org/pdf/1703.08581.pdf) 173 | 174 | Ron J. Weiss, Jan Chorowski, Navdeep Jaitly, Yonghui Wu, Zhifeng Chen 175 | 176 | * [Related paper to read](http://aclweb.org/anthology/E17-2076) 177 | 178 | **Tuesday 6 June 2017** 179 | 180 | [Covonlutional Sequence to Sequence Learning](https://arxiv.org/abs/1705.03122) 181 | 182 | Jonas Gehring, Michael Auli, David Grangier, Denis Yarats, Yann N. Dauphin 183 | 184 | **Tuesday 30 May 2017** 185 | 186 | [Program Induction by Rationale Generation:Learning to Solve and Explain Algebraic Word Problems](https://arxiv.org/abs/1705.04146) 187 | 188 | Wang Ling, Dani Yogatama, Chris Dyer, Phil Blunsom 189 | 190 | **Tuesday 9 May 2017** 191 | 192 | Chatterjee et al.: [Online Automatic Post-editing for MT in a Multi-Domain Translation Environment](http://aclweb.org/anthology/E17-1050) 193 | 194 | **Tuesday 6 May 2017** 195 | 196 | [Convolutional Sequence to Sequence Learning](https://arxiv.org/abs/1705.03122) 197 | 198 | Jonas Gehring, Michael Auli, David Grangier, Denis Yarats, Yann N. Dauphin 199 | 200 | **Tuesday 2 May 2017** 201 | 202 | [Coarse-to-Fine Question Answering for Long Documents](http:// http://homes.cs.washington.edu/~eunsol/papers/acl17eunsol.pdf) 203 | 204 | **Tuesday 25 April 2017** 205 | 206 | [Re-evaluating Automatic Metrics for Image Captioning](https://arxiv.org/pdf/1612.07600.pdf) 207 | 208 | Mert Kilickaya, Aykut Erdem, Nazli Ikizler-Cinbis, Erkut Erdem 209 | 210 | **Tuesday 18 April 2017** 211 | 212 | [Neural Tree Indexers](https://www.aclweb.org/anthology/E/E17/E17-1002.pdf), EACL2017 213 | 214 | **Tuesday 11 April 2017** 215 | 216 | EACL Recap 217 | 218 | **Tuesday 4 April 2017** 219 | 220 | [Shakir Mohammed's deep learning overview](http://blog.shakirm.com/2015/01/a-statistical-view-of-deep-learning-i-recursive-glms/ ) 221 | 222 | **Tuesday 28 March 2017** 223 | 224 | [Abstractive Text Summarization using Sequence-to-sequence RNNs and Beyond](http://www.aclweb.org/anthology/K16-1028) 225 | 226 | **Tuesday 21 March 2017** 227 | 228 | [Unsupervised AMR-Dependency Parse Alignment](http://verbs.colorado.edu/~wech5560/paper/EACL17.pdf) 229 | 230 | **Tuesday 14 March 2017** 231 | 232 | Kim et al. (2016): [Examples are not Enough, Learn to Criticize!](http://(http://people.csail.mit.edu/beenkim/papers/KIM2016NIPS_MMD.pdf)) Criticism for Interpretability, NIPS 2016 233 | 234 | **Tuesday 7 March 2017** 235 | 236 | [Latent Variable Dialogue Models and their Diversity](https://arxiv.org/pdf/1702.05962.pdf) 237 | 238 | Kris Cao and Stephen Clark 239 | 240 | **Tuesday 28 February 2017** 241 | 242 | [Zhang et al. EACL2017](https://arxiv.org/pdf/1606.01280.pdf) 243 | 244 | **Tuesday 21 February 2017** 245 | 246 | [Structured Attention Networks](https://arxiv.org/pdf/1702.00887.pdf) 247 | 248 | **Tuesday 14 February 2017** 249 | 250 | [CORE: Context-Aware Open Relation Extraction with Factorization Machines](http://www.emnlp2015.org/proceedings/EMNLP/pdf/EMNLP204.pdf) 251 | 252 | by Fabio Petroni, Luciano Del Corro and Rainer Gemulla 253 | 254 | **Tuesday 7 February 2017** 255 | 256 | [Adversarial Training Methods for Semi-Supervised Text Classification](https://arxiv.org/pdf/1605.07725v2.pdf) 257 | 258 | Takeru Miyato, Andrew, M.Dai, Ian Goodfellow 259 | 260 | * [ICLR paper by Goodfellow et.al](https://arxiv.org/pdf/1412.6572v3.pdf) 261 | 262 | **Tuesday 31 January 2017** 263 | 264 | [Learning to Prune: Exploring the Frontier of Fast and Accurate Parsing](https://timvieira.github.io/doc/2016-tacl-pruning.pdf) 265 | 266 | Tim Vieira and Jason Eisner 267 | 268 | * [Slides](https://timvieira.github.io/doc/2016-tacl-pruning-slides.pdf) 269 | 270 | **Tuesday 24 January 2017** 271 | 272 | [Matching Networks for One Shot Learning](https://papers.nips.cc/paper/6385-matching-networks-for-one-shot-learning.pdf) 273 | 274 | Oriol Vinyals, Charles Blundell, Tim Lillicrap, Koray Kavukcuoglu, Daan Wierstra 275 | 276 | **Tuesday 17 January 2017** 277 | 278 | [Learning Structured Predictors from Bandit Feedback for Interactive NLP](http://www.stefanriezler.com/PAPERS/ACL2016.pdf). In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (ACL). Berlin, Germany 279 | 280 | Artem Sokolov, Julia Kreutzer, Christopher Lo, Stefan Riezler 281 | 282 | **Tuesday 13 December 2016** 283 | 284 | [Optimization and Sampling for NLP from a Unified Viewpoint](http://aclweb.org/anthology/W/W12/W12-6106.pdf) 285 | 286 | Marc Dymetman, Guillaume Bouchard, Simon Carter 287 | 288 | **Tuesday 6 December 2016** 289 | 290 | [Matrix Completion has No Spurious Local Minimum](https://arxiv.org/abs/1605.07272) 291 | 292 | Rong Ge, Jason D. Lee, Tengyu Ma 293 | 294 | **Tuesday 29 November 2016** 295 | 296 | [Compositional Semantic Parsing on Semi-Structured Tables](https://arxiv.org/pdf/1508.00305v1.pdf)  297 | Panupong Pasupat and Percy Liang 298 | 299 | **Tuesday 22 November 2016** 300 | 301 | [Minimum Risk Training for Neural Machine Translation](https://www.aclweb.org/anthology/P/P16/P16-1159.pdf)  302 | Shiqi Shen, Yong Cheng, Zhougjun He, Wei He, Hua Wu, Maosong Sun, Yang Liu 303 | 304 | **Tuesday 15 November 2016** 305 | 306 | [Generation from Abstract Meaning Representation using Tree Transducers](https://www.cs.cmu.edu/~jgc/publication/flanigantree.pdf)  307 | Jeffrey Flanigan, Chris Dyer, Noah A. Smith and Jaime Carbonell 308 | 309 | **Tuesday 1 November 2016** 310 | 311 | [Visual Representations for Topic Understanding and Their Effects on Manually Generated Labels](https://staffwww.dcs.shef.ac.uk/people/M.Stevenson/campus_only/Evaluating_Visual_Representations_for_Topic_Understanding.pdf) Transactions of the Association for Computational Linguistics, 2016.  312 | Alison Smith, Tak Yeon Lee, Forough Poursabzi-Sangdeh, Leah Findlater, Jordan Boyd-Graber, and Niklas Elmqvist 313 | 314 | **Tuesday 25 October 2016** 315 | 316 | [Learning to Search Better than your Teacher](https://arxiv.org/pdf/1502.02206.pdf) 317 | 318 | [Talk](https://www.cs.umd.edu/media/2015/12/video/17235-daume-stuff-i-did-spring-while-not-replying-email)  319 | Chang et al. ICML 2015 320 | 321 | **Tuesday 11 October 2016** 322 | 323 | [A Thorough Examination of the CNN/Daily Mail Reading Comprehension Task](https://arxiv.org/pdf/1606.02858v2.pdf)  324 | Danqi Chen, Jason Bolton, Christopher D. Manning 325 | 326 | **Tuesday 4 October 2016** 327 | 328 | [Ultradense Word Embeddings by Orthogonal Transformation](http://arxiv.org/abs/1602.07572)  329 | Sascha Rothe, Sebastian Ebert, Hinrich Schütze 330 | 331 | **Tuesday 7 June 2016** 332 | 333 | [Not All Character _N_-grams Are Created Equal: A Study in Authorship Attribution.](http://www.aclweb.org/anthology/N15-1010.pdf)  334 | Upendra Sapkota, Steven Bethard, Manuel Montes-y-Gómez & Thamar Solorio (2015) 335 | 336 | **Tuesday 31 May 2016** 337 | 338 | [Relation extraction with matrix factorization and universal schemas.](http://www.riedelcastro.org/publications/papers/riedel13relation.pdf) 339 | 340 | Riedel, S., Yao, L., McCallum, A., & Marlin, B. M. (2013) 341 | 342 | **Tuesday 10 May 2016** 343 | 344 | [Training Deterministic Parsers with Non-Deterministic Oracles, TACL](https://www.aclweb.org/anthology/Q/Q13/Q13-1033.pdf) 345 | 346 | [slides](https://www.rocq.inria.fr/alpage-wiki/tiki-download_wiki_attachment.php?attId=3)  347 | Goldberg, Y. and Nivre, J. (2013) 348 | 349 | **Tuesday 3 May 2016** 350 | 351 | [A New Corpus and Imitation Learning Framework for Context-Dependent Semantic Parsing](http://www.aclweb.org/anthology/Q14-1042.pdf)  352 | Vlachos, A. and Clark, S. 353 | 354 | **Tuesday 22 April 2016** 355 | 356 | [Sequence Level Training with recurrent Neural Networks](http://arxiv.org/pdf/1511.06732v5.pdf)  357 | Marc'Aurelio Ranzato, Sumit Chopra, Michael Auli, Wojciech Zaremba 358 | 359 | **Tuesday 22 March 2016** 360 | 361 | ["Distributed Representation of Sentences and Documents"](https://cs.stanford.edu/~quocle/paragraph_vector.pdf)  362 | Quoc Le and Tomas Mikolov 363 | 364 | **Tuesday 8 March 2016** 365 | 366 | [AutoExtend: Extending Word Embeddings to Embeddings for Synsets and Lexemes](http://www.aclweb.org/anthology/P15-1173)  367 | Sascha Rothe; Hinrich Schütze. ACL2015 ([best student paper](http://www.aclweb.org/anthology/P/P15/P15-1173.bib)) 368 | 369 | **Tuesday 23 February 2016** 370 | 371 | [From Word Embeddings To Document Distances](http://jmlr.org/proceedings/papers/v37/kusnerb15.pdf)  372 | Kusner et al. 373 | 374 | **Tuesday 16 February 2016** 375 | 376 | ["Target-Dependent Twitter Sentiment Classification with Rich Automatic Features"](http://ijcai.org/papers15/Papers/IJCAI15-194.pdf) 377 | 378 | **Tuesday 9 February 2016** 379 | 380 | ["Evaluation methods for unsupervised word embeddings"](http://www.emnlp2015.org/proceedings/EMNLP/pdf/EMNLP036.pdf) 381 | 382 | **Tuesday 25 January 2016** 383 | 384 | [Multi-Perspective Sentence Similarity Modeling with Convolutional Neural Networks](http://ttic.uchicago.edu/~kgimpel/papers/he+etal.emnlp15.pdf)  385 | Hua He, Kevin Gimpel, and Jimmy Lin. EMNLP2015 386 | 387 | **Tuesday 19 January 2016** 388 | 389 | [Multilingual Image Description with Neural Sequence Models](http://arxiv.org/abs/1510.04709) 390 | 391 | **Tuesday 12 January 2016** 392 | 393 | ["Improving Distributional Similarity with Lessons Learned from Word Embeddings"](https://levyomer.wordpress.com/2015/03/30/improving-distributional-similarity-with-lessons-learned-from-word-embeddings/) 394 | 395 | **Tuesday 8 December 2015** 396 | 397 | [Using Discourse Structure Improves Machine Translation Evaluation](http://www.aclweb.org/anthology/P/P14/P14-1065.pdf).  398 | F Guzmán, S Joty, L Màrquez, P Nakov 399 | 400 | And here are the author's [slides](http://alt.qcri.org/~guzmanhe/media/ACL2014-Guzman.pdf) 401 | 402 | **Tuesday 1 December 2015** 403 | 404 | [Practical Bayesian Optimization of Machine Learning Algorithms Advances in Neural Information Processing Systems](https://hips.seas.harvard.edu/files/snoek-bayesopt-nips-2012.pdf), 2012  405 | Snoek, J.; Larochelle, H. & Adams, R. P. 406 | 407 | Related presentations/lecture slides: 408 | 409 | [http://becs.aalto.fi/en/research/bayes/courses/4613/Vik_Kamath_Presentation.pdf](http://becs.aalto.fi/en/research/bayes/courses/4613/Vik_Kamath_Presentation.pdf) 410 | 411 | [http://drona.csa.iisc.ernet.in/~indous/Lectures-2014/slides/jasper.pdf](http://drona.csa.iisc.ernet.in/~indous/Lectures-2014/slides/jasper.pdf) 412 | 413 | [Related Video](https://www.youtube.com/watch?v=a79klpzaPgY) 414 | 415 | My reading group [presentation slides](https://drive.google.com/file/d/0BwXi9bHZYCleVktIOTAxRE9RcG8/view?usp=sharing) 416 | 417 | **Tuesday 24 November 2015** 418 | 419 | [Improved Semantic Representations From Tree-Structured Long Short-Term Memory Networks](http://arxiv.org/abs/1503.00075) ACL 2015  420 | LSTMs? Kai Sheng Tai, Richard Socher, Christopher D. Manning 421 | 422 | [http://www.wildml.com/2015/09/recurrent-neural-networks-tutorial-part-2-implementing-a-language-model-rnn-with-python-numpy-and-theano/](http://www.wildml.com/2015/09/recurrent-neural-networks-tutorial-part-2-implementing-a-language-model-rnn-with-python-numpy-and-theano/) 423 | 424 | [http://www.wildml.com/2015/10/recurrent-neural-networks-tutorial-part-3-backpropagation-through-time-and-vanishing-gradients/](http://www.wildml.com/2015/10/recurrent-neural-networks-tutorial-part-3-backpropagation-through-time-and-vanishing-gradients/) 425 | 426 | [http://colah.github.io/posts/2015-08-Understanding-LSTMs/](http://colah.github.io/posts/2015-08-Understanding-LSTMs/) 427 | 428 | Additional resource about LSTM: ["Anyone Can Learn To Code an LSTM-RNN in Python"](http://iamtrask.github.io/2015/11/15/anyone-can-code-lstm/) 429 | 430 | **Tuesday 17 November 2015** 431 | 432 | [RNNs/LSTMs](http://www-cs.stanford.edu/~quocle/tutorial2.pdf) [ConvNets](http://www.wildml.com/2015/11/understanding-convolutional-neural-networks-for-nlp/) 433 | 434 | More details on auto encoders for unsupervised pre-training: 435 | 436 | [http://deeplearning.stanford.edu/wiki/index.php/Autoencoders_and_Sparsity](http://deeplearning.stanford.edu/wiki/index.php/Autoencoders_and_Sparsity) 437 | 438 | [http://www.jmlr.org/papers/volume11/erhan10a/erhan10a.pdf](http://www.jmlr.org/papers/volume11/erhan10a/erhan10a.pdf) 439 | 440 | [http://www.slideshare.net/billlangjun/simple-introduction-to-autoencoder](http://www.slideshare.net/billlangjun/simple-introduction-to-autoencoder) 441 | 442 | **Tuesday 10 November 2015** 443 | 444 | [Multi-Metric Optimization Using Ensemble Tuning](http://www.aclweb.org/anthology/N13-1115). NAACL2013. [Video](http://techtalks.tv/talks/multi-metric-optimization-using-ensemble-tuning/58498/)  445 | Baskaran Sankaran, Anoop Sarkar and Kevin Duh 446 | 447 | **Tuesday 3 November 2015** 448 | 449 | [NN tutorials by Quoc Le](http://www-cs.stanford.edu/~quocle/tutorial1.pdf) 450 | 451 | Josiah's [slides](https://drive.google.com/file/d/0B99oZRtWIpwlUTVGYjlvWERKTWc/view?usp=sharing) 452 | 453 | Other resources: 454 | 455 | [Andrej Karpathy's notes](http://cs231n.github.io/) 456 | 457 | [Different objective functions, multiclass problems](http://cs231n.github.io/linear-classify/) 458 | 459 | [Gradient descent](http://cs231n.github.io/optimization-1/) 460 | 461 | [Backpropagation](http://cs231n.github.io/optimization-2/) 462 | 463 | [Discussion about different activation functions](http://cs231n.github.io/neural-networks-1/) 464 | 465 | **Tuesday 27 October 2015** 466 | 467 | [Three blog posts introducing RNNs for language modelling in equations and code](http://www.wildml.com/2015/09/recurrent-neural-networks-tutorial-part-1-introduction-to-rnns/) 468 | 469 | might help to read this [NLP primer](http://u.cs.biu.ac.il/~yogo/nnlp.pdf) 470 | 471 | Additional material: 472 | [a thorough explanation of back propagation](http://neuralnetworksanddeeplearning.com/chap2.html) 473 | 474 | **Tuesday 20 October 2015** 475 | 476 | [Teaching Machines to Read and Comprehend](http://arxiv.org/abs/1506.03340). NIPS 2015.  477 | Karl Moritz Hermann, Tomáš Kociský, Edward Grefenstette, Lasse Espeholt, Will Kay, Mustafa Suleyman, Phil Blunsom 478 | 479 | [Slides](http://lxmls.it.pt/2015/lxmls15.pdf) (presented at LXMLS) 480 | 481 | _Background reading_: 482 | 483 | [Understanding LSTMs](http://colah.github.io/posts/2015-08-Understanding-LSTMs/) 484 | 485 | NAACL 2013 Tutorial ["Deep Learning without Magic"](http://nlp.stanford.edu/courses/NAACL2013/NAACL2013-Socher-Manning-DeepLearning.pdf) 486 | 487 | EMNLP 2014 Tutorial ["Embedding Methods for NLP"](http://emnlp2014.org/tutorials/8_notes.pdf) 488 | 489 | Related Work: 490 | 491 | [Entailment with Neural Attention](http://arxiv.org/abs/1509.06664) (better description of attention models than in the NIPS paper in my opinion) 492 | 493 | [Memory Networks](https://cs224d.stanford.edu/lectures/CS224d-Lecture12.pdf) 494 | 495 | **Tuesday 13 October 2015** 496 | 497 | [A large annotated corpus for learning natural language inference](http://nlp.stanford.edu/pubs/snli_paper.pdf). Proceedings of EMNLP 2015.  498 | Samuel R. Bowman, Gabor Angeli, Christopher Potts, and Christopher D. Manning 499 | 500 | Should compare this to work on (multilingual) textual similarity 501 | -------------------------------------------------------------------------------- /conferences_recap/emnlp2018_recap.md: -------------------------------------------------------------------------------- 1 | # EMNLP 2018 recap 2 | 3 | ### Carol 4 | 5 | * Maruf et al. (2018): Contextual Neural Model for Translating Bilingual Multi-Speaker Conversations (WMT 2018) [PDF](http://aclweb.org/anthology/W18-6311) 6 | * Kann et al. (2018): Sentence-Level Fluency Evaluation: References Help, But Can Be Spared! (CONLL 2018) [PDF](http://aclweb.org/anthology/K18-1031) 7 | 8 | *Other papers* 9 | * Stojanovski and Fraser (2018): Coreference and Coherence in Neural Machine Translation: A Study Using Oracle Experiments (WMT 2018) [PDF](http://www.statmt.org/wmt18/pdf/WMT006.pdf) 10 | * Post (2018): A Call for Clarity in Reporting BLEU Scores (WMT 2018) [PDF](http://aclweb.org/anthology/W18-6319) 11 | 12 | ### Hardy 13 | 14 | * Shashi et al. (2018): Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization [PDF](http://aclweb.org/anthology/D18-1206) 15 | * Kedzie et al. (2018): Content Selection in Deep Learning Models of Summarization [PDF](http://aclweb.org/anthology/D18-1208) 16 | * Kryściński et al. (2018): Improving Abstraction in Text Summarization [PDF](http://aclweb.org/anthology/D18-1207) 17 | 18 | ### Abiola 19 | 20 | * Kiros and Chan (2018): InferLite: Simple Universal Sentence Representations from Natural Language Inference Data [PDF](http://aclweb.org/anthology/D18-1524) 21 | * Yang (2018): Convolutional Neural Networks with Recurrent Neural Filters [PDF](http://aclweb.org/anthology/D18-1109) 22 | 23 | 24 | -------------------------------------------------------------------------------- /docs/_config.yml: -------------------------------------------------------------------------------- 1 | theme: jekyll-theme-hacker 2 | -------------------------------------------------------------------------------- /docs/index.html: -------------------------------------------------------------------------------- 1 |

NLP Reading Group

2 |

The target audience is all the members of the NLP group and other possible interested participants.

3 |

The meeting takes place weekly during term for one hour usually on Wednesdays, 12-13:30pm. See below a list of the upcoming and past meetings.

4 |

The meetings of the group are informal and no necessary preparation will be required with the exception of the moderator reading the current paper and the rest having at least a brief overview of it.

5 |

A list of past meetings before 2018/19 can be found here.

6 |

2018-2019

7 |

Autumn Semester

8 |

Upcoming Meetings

9 | 69 |
70 |

Past Meetings

71 | --------------------------------------------------------------------------------