└── readme.md /readme.md: -------------------------------------------------------------------------------- 1 | # 论文仓库 2 | 3 | Created: Sep 4, 2020 4:31 PM 4 | 5 | 不断完善中，欢迎补充Slide地址~ 6 | 7 | 8 | 初衷是对不久前的KBQA调研做一个总结，未来可能会整理成一个多个KG方向资源的仓库 9 | 10 | # **KBQA-paper** 11 | 12 | ## **KBQA系统** 13 | 14 | ### **分阶段生成** 15 | 16 | 1. Semantic parsing via staged query graph generation: Question answering with knowledge base(ACL2015 MS Wen-tau Yih) SP方法开山之作 17 | 2. Formal Query Building with Query Structure Prediction for Complex Question Answering over Knowledge Base(AAAI2020 东南大学陈永锐) 18 | 3. SPARQA: skeleton-based semantic parsing for complex questions over knowledge bases（AAAI 2020 南京大学） 19 | 4. Leveraging Frequent Query Substructures to Generate Formal Queries for Complex Question Answering （EMNLP 2019 南京大学） 20 | 5. Query Graph Generation for Answering Multi-hop Complex Questions from Knowledge Bases(ACL2020 SMU 蓝韵诗) 21 | 6. Knowledge base question answering via encoding of complex query graphs(EMNLP2018 上交) 22 | 7. A state-transition framework to answer complex questions over knowledge base(EMNLP2018 北大胡森) 23 | 8. Answering Complex Questions by Combining Information from Curated and Extracted Knowledge Bases（ACL2020） 24 | 25 | ### **IR** 26 | 1. Open Domain Question Answering Using Early Fusion of Knowledge Bases and Text（EMNLP 2018 Googel Sun haitian) 27 | 2. PullNet: Open domain question answering with iterative retrieval on knowledge bases and text(EMNLP 2019 Googel Sun haitian) 28 | 3. Improving Multi-hop Knowledge Base Question Answering by Learning Intermediate Supervision Signals（WSDM2021 SMU 蓝韵诗） 29 | 30 | ### **其他方法** 31 | 1. Case-based Reasoning for Natural Language Queries over Knowledge Bases （Google 2021） 32 | 2. TransferNet: An Effective and Transparent Framework for Multi-hop Question Answering over Relation Graph(2021 清华史佳欣) 33 | 34 | ### **模板** 35 | 36 | 1. KBQA: Learning question answering over QA corpora and knowledge bases（PVLDB2018 复旦崔万云） 37 | 2. Question answering over knowledge graphs: Question understanding via template decomposition(PVLDB2018 复旦郑卫国) 38 | 3. Leveraging frequent query substructures to generate formal queries for complex question answering(EMNLP2019 南京大学) 39 | 4. Automated template generation for question answering over knowledge graphs(WWW2017 马普所) 40 | 41 | ### **Seq2Seq** 42 | 43 | 1. KQA Pro: A Large-Scale Dataset with Interpretable Programs and Accurate SPARQLs for Complex Question Answering over Knowledge Base(ACL2022) 44 | 2. SKG: Unifying and Multi-Tasking Structured Knowledge Grounding with Text-to-Text Language Models 45 | 3. Unseen Entity Handling in Complex Question Answering over Knowledge Base via Language Generation(EMNLP2021) 46 | 4. RNG-KBQA: Generation Augmented Iterative Ranking for Knowledge Base Question Answering(ACL2022) 47 | 5. Constraint-based question answering with knowledge graph(COLING2016) 48 | 6. Sequence-based structured prediction for semantic parsing(ACL2016) 49 | 7. Language to logical form with neural attention(ACL2016) 50 | 8. Neural symbolic machines: Learning semantic parsers on freebase with weak supervision(ACL2017) 51 | 9. Learning a neural semantic parser from user feedback(ACL2017) 52 | 10. Coarse-to-fine decoding for neural semantic parsing(ACL2018) 53 | 11. A syntactic neural model for general-purpose code generation（ACL2017） 54 | 12. Sequence-to-action: End-to-end semantic graph generation for semantic parsing(ACL2018) 55 | 13. Language to logical form with neural attention（ACL2016） 56 | 57 | ### **复杂问题分解** 58 | 59 | 1. The web as a knowledge-base for answering complex questions（NAACL2018）CWQ数据集 基于搜索引擎+RC回答子问题 [GitHub](https://github.com/alontalmor/WebAsKB) 60 | 2. EDG-based Question Decomposition for ComplexQuestion Answering over Knowledge Bases(ISWC2021 南京大学) 61 | 3. Complex question decomposition for semantic parsing（ACL 2020 国防科大） 62 | 4. SPARQA: skeleton-based semantic parsing for complex questions over knowledge bases（AAAI 2020 南京大学） 63 | 5. BREAK it down: A question understanding benchmark(TACL 2019 AI2) 64 | 6. Text modular networks: learning to decompose tasks in the language of existing models(NAACL 2021 AI2) 65 | 7. KQA Pro: A Large-Scale Dataset with Interpretable Programs and Accurate SPARQLs for Complex Question Answering over Knowledge Base（WWW2021 清华史佳欣） 66 | 8. Multi-hop reading comprehension through question decomposition and rescoring(ACL 2019 RC的分解) 67 | 9. Unsupervised question decomposition for question answering （EMNLP 2020 FBAI） 68 | 10. Enhancing key-value memory neural networks for knowledge based question answering（NAACL2019） 69 | 11. Did aristotle use a laptop? A question answering benchmark with implicit reasoning strategies(TACL2020) QDMR的结构 70 | 12. MuSiQue: Multi-hop Questions via Single-hop Question Composition(TACL 2022) QDMR的结构 71 | 13. Break, Perturb, Build : Automatic Perturbation of Reasoning Paths Through Question Decomposition(TACL2022) QDMR的结构 72 | 14. SPARQLing Database Queries from Intermediate Question Decompositions(EMNLP2021) QDMR的结构 text2sql 73 | 15. Weakly Supervised Mapping of Natural Language to SQL through Question Decomposition QDMR的结构 text2sql 74 | 75 | 76 | 77 | ## **综述** 78 | 79 | 1. Survey on challenges of Question Answering in the Semantic Web（Semantic Web2017） 80 | 2. A Survey on Complex Knowledge Base Question Answering: Methods, Challenges and Solutions （IJCAI 2021 SMU 蓝韵诗） 81 | 3. Complex Knowledge Base Question Answering:A Survey(蓝韵诗 2021) 82 | 4. Core Techniques of Question Answering Systems over Knowledge Bases : a Survey（2017） 83 | 5. A Survey on Complex Question Answering over Knowledge Base: Recent Advances and Challenges（2020 阿里巴巴），中文翻译：[https://zhuanlan.zhihu.com/p/134090164](https://zhuanlan.zhihu.com/p/134090164) 84 | 6. Introduction to Neural Network based Approaches for Question Answering over Knowledge Graphs(2019) 85 | 7. Question Answering over Curated and Open Web Sources (SIGIR2020 Tutorial) [Slide](http://people.mpi-inf.mpg.de/~rsaharo/sigir20slides.pdf) 86 | 87 | ## **子任务** 88 | 89 | ### **Relation Linking/Detection** 90 | 91 | 1. Old is Gold: Linguistic Driven Approach for Entity and Relation Linking of Short Text（NAACL 2019）([Falcon系统](https://labs.tib.eu/falcon/) [GitHub](https://github.com/AhmadSakor/falcon)) 92 | 2. EARL: Joint entity and relation linking for question answering over knowledge graphs （ISWC 2018） 93 | 3. Leveraging semantic parsing for relation linking over knowledge bases (ISWC2020 IBM Watson) (SOTA) 94 | 4. Generative Relation Linking for Question Answering over Knowledge Bases(ISWC2021 IBM Watson) 95 | 5. FALCON 2.0: An Entity and Relation Linking Tool over Wikidata （ISWC 2019）（[Falcon2.0](https://labs.tib.eu/falcon/falcon2/) [GitHub](https://github.com/SDM-TIB/Falcon2.0)）（开源系统SOTA） 96 | 6. Scalable knowledge graph construction over text using deep learning based predicate mapping （WWW2019） 97 | 7. Learning Representation Mapping for Relation Detection in Knowledge Base Question Answering(ACL2019 南京大学) 98 | 8. PATTY: A taxonomy of relational patterns with semantic types （EMNLP 2012 马普所） 99 | 9. Towards Combinational Relation Linking over Knowledge Graphs （复旦） 100 | 10. LEVERAGING SEMANTIC PARSING FOR RELATION LINKING OVER KNOWLEDGE BASES(ISWC2021) 101 | 11. LNN-EL: A Neuro-Symbolic Approach to Short-text Entity Linking(ACL21) 102 | 12. EntQA: Entity Linking as Question Answering(ICLR2022) 103 | 13. A Semantics-aware Transformer Model of Relation Linking for Knowledge Base Question Answering(ACL21 IBM, QALD9,LC1,2,simplequestion) 104 | 105 | 106 | ### **Entity Linking** 107 | 108 | 1. LNN-EL: A Neuro-Symbolic Approach to Short-text Entity Linking （ACL2021 IBM Watson） 109 | 2. Autoregressive entity retrieval(ICLR 2021 FBAI) 110 | 3. Efficient One-Pass End-to-End Entity Linking for Questions(EMNLP 2020 FBAI) 111 | 4. Scalable Zero-shot Entity Linking with Dense Entity Retrieval(EMNLP 2020 FBAI) 112 | 5. PNEL: Pointer Network based End-To-End Entity Linking over Knowledge Graphs (ISWC 2020) 113 | 6. KBPearl: A knowledge base population system supported by joint entity and relation linking(PVLDB 2020) 114 | 115 | 116 | ## **Futrue direction** 117 | 118 | ### **可解释性** 119 | 120 | ### **鲁棒性** 121 | 122 | ### **多源知识融合** 123 | 124 | ### **对话QA** 125 | 126 | ## **数据集** 127 | 128 | 1. **ComplexQuestion**: Constraint-based question answering with knowledge graph(COLING2016) 129 | 微软亚研院周明组提出 130 | 131 | 论文：[Constraint-Based Question Answering with Knowledge Graph](https://www.aclweb.org/anthology/C16-1236/) 132 | 133 | 数据集：[https://github.com/JunweiBao/MulCQA/tree/ComplexQuestions](https://github.com/JunweiBao/MulCQA/tree/ComplexQuestions) 134 | 135 | 简介：2100条（1300/800） 136 | 137 | 2. **ComplexWebQuestion**: 138 | 139 | The web as a knowledge-base for answering complex questions(NAACL2018) 140 | 141 | 子问题的回答基于搜索引擎上的RC 142 | 3. **WebQuestion** [train](http://nlp.stanford.edu/static/software/sempre/release-emnlp2013/lib/data/webquestions/dataset_11/webquestions.examples.train.json.bz2) [test](http://nlp.stanford.edu/static/software/sempre/release-emnlp2013/lib/data/webquestions/dataset_11/webquestions.examples.test.json.bz2) 143 | 144 | 论文：Semantic Parsing on Freebase from Question-Answer Pairs（EMNLP2013斯坦福） 145 | 146 | 简介：5810条（3778/2032） 斯坦福大学，根据Google Suggest API构建 147 | 4. **WebQuestionSP**: 148 | 149 | Semantic Parsing via Staged Query Graph Generation: Question Answering with Knowledge Base（ACL 2015） 150 | 151 | 微软yih-wen tau 152 | 5. **SimpleQuestion** 153 | 154 | 论文：[Large-scale Simple Question Answering with Memory Networks](https://arxiv.org/pdf/1506.02075.pdf)（2015） 155 | 156 | 数据集：[https://research.fb.com/downloads/babi/](https://research.fb.com/downloads/babi/) 157 | 158 | 简介：108442条（75910/10845/21687）Freebase 全是一个三元组就可以回答的问题 159 | 160 | 6. **Free917** 161 | [train](http://nlp.stanford.edu/static/software/sempre/release-emnlp2013/data/free917.train.examples.canonicalized.json.bz2) [test](http://nlp.stanford.edu/static/software/sempre/release-emnlp2013/data/free917.test.examples.canonicalized.json.bz2) 162 | 163 | 简介：917条 Freebase 164 | 7. **QALD series**: 165 | 资源地址：[https://github.com/ag-sc/QALD](https://github.com/ag-sc/QALD) 166 | 167 | 论文 168 | 169 | QALD-7: [https://svn.aksw.org/papers/2017/ESWC_2017_QALD/public.pdf](https://svn.aksw.org/papers/2017/ESWC_2017_QALD/public.pdf) 170 | 171 | QALD-8: [http://ceur-ws.org/Vol-2241/paper-05.pdf](http://ceur-ws.org/Vol-2241/paper-05.pdf) 172 | 173 | QALD-9: [http://ceur-ws.org/Vol-2241/paper-06.pdf](http://ceur-ws.org/Vol-2241/paper-06.pdf) 174 | 175 | 评测地址 176 | 177 | QALD-7: [http://gerbil-qa.aksw.org/gerbil/experiment?id=201706300001](http://gerbil-qa.aksw.org/gerbil/experiment?id=201706300001) 178 | 179 | QALD-8: [http://gerbil-qa.aksw.org/gerbil/experiment?id=201710220000](http://gerbil-qa.aksw.org/gerbil/experiment?id=201710220000) 180 | 181 | QALD-9: [http://gerbil-qa.aksw.org/gerbil/experiment?id=201810080002](http://gerbil-qa.aksw.org/gerbil/experiment?id=201810080002) 182 | 183 | 8. **LC-QuAD**: [http://lc-quad.sda.tech/lcquad1.0.html](http://lc-quad.sda.tech/lcquad1.0.html) 184 | 9. **LC-QuAD 2.0**: [http://lc-quad.sda.tech/](http://lc-quad.sda.tech/) 185 | ``` 186 | Question Answering over Knowledge Graphs with Neural Machine Translation and Entity Linking 187 | ``` 188 | 10. **KQA Pro** 189 | 190 | [KQA Pro: A Large-Scale Dataset with Interpretable Programs and Accurate SPARQLs for Complex Question Answering over Knowledge Base(WWW 2021 清华)](https://arxiv.org/abs/2007.03875) 191 | 192 | [github](https://github.com/shijx12/KQAPro_Baselines) 193 | 194 | [homepage](http://thukeg.gitee.io/kqa-pro/) 195 | 196 | wikidata，120K有SPARQL标注，KB子集 197 | 11. **CRONKGQA** 198 | 199 | [Question Answering over Temporal Knowledge Graphs（ACL2021)](https://aclanthology.org/2021.acl-long.520/) 200 | 201 | [github](https://github.com/apoorvumang/CronKGQA) 202 | 203 | 时间问答数据集，wikidata，410K,KB子集 204 | 12. **CoQA** 205 | 13. **GraphQuestion** 206 | 207 | 论文：[On Generating Characteristic-rich Question Sets for QA Evaluation](https://www.aclweb.org/anthology/D16-1054/)（2016） 208 | 209 | 数据集：[https://github.com/ysu1989/GraphQuestions](https://github.com/ysu1989/GraphQuestions) 210 | 211 | 简介：Freebase 212 | 213 | 12. **30M Factoid Question** 214 | 215 | 论文：[Generating Factoid QuestionsWith Recurrent Neural Networks: The 30M Factoid Question-Answer Corpus](https://arxiv.org/pdf/1603.06807.pdf) 216 | 217 | 数据集：[https://academictorrents.com/details/973fb709bdb9db6066213bbc5529482a190098ce](https://academictorrents.com/details/973fb709bdb9db6066213bbc5529482a190098ce) 218 | 219 | 简介：30M条问答对 220 | 221 | 13. **MetaQA 222 | 数据集：**[https://drive.google.com/drive/folders/0B-36Uca2AvwhTWVFSUZqRXVtbUE](https://drive.google.com/drive/folders/0B-36Uca2AvwhTWVFSUZqRXVtbUE) 223 | 14. **CSQA** 224 | 15. **ReClor** 225 | 论文: ReClor: A Reading Comprehension Dataset Requiring Logical Reasoning(ICLR2020) 226 | 项目地址：[http://whyu.me/reclor/](http://whyu.me/reclor/) 227 | 16. **CLEVRER** 228 | 论文: [CLEVRER: CoLlision Events for Video REpresentation and Reasoning(ICLR2020)](https://arxiv.org/pdf/1910.01442.pdf) 229 | 项目地址: [http://clevrer.csail.mit.edu/](http://clevrer.csail.mit.edu/) 230 | 17. **CommonsenseQA** 231 | 论文: Cosmos QA: Machine Reading Comprehension with Contextual Commonsense Reasoning 232 | 特拉维夫大学常识问答任务,SOTA(0.9179) 233 | 项目地址: [https://www.tau-nlp.org/commonsenseqa](https://www.tau-nlp.org/commonsenseqa) 234 | 18. **Hotpot QA** 235 | 19. **TriviaQA** 阅读理解数据集 236 | 论文:TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension 237 | 20. **Mutual** 对话数据集 238 | 西湖大学张岳组 239 | **论文地址：**[MuTual: A Dataset for Multi-Turn Dialogue Reasoning](http://arxiv.org/abs/2004.04494) 240 | **github地址：**[https://github.com/Nealcly/MuTu](https://link.zhihu.com/?target=https%3A//github.com/Nealcly/MuTual) 241 | 242 | ## 优秀开源工具 243 | 244 | 1. SEMPRE: Semantic Parsing with Execution 斯坦福 245 | [项目地址](https://nlp.stanford.edu/software/sempre/) [tutorial](https://github.com/percyliang/sempre/blob/master/TUTORIAL.md) [GitHub](https://github.com/percyliang/sempre) 246 | - **简介：**NL转换成逻辑形式，支持lambda calculus，lambda DCS等 247 | 2. Stanza: A Python NLP Library for Many Human Languages斯坦福官方NLP工具包 248 | ACL2020 demo 249 | 250 | ## 其他开源工具 251 | 252 | 1. 中文近义词 [https://github.com/chatopera/Synonyms](https://github.com/chatopera/Synonyms) 253 | 254 | ## 讲习班 Slide 255 | 256 | 学习讲习班大佬们的总结，可以迅速了解一个领域 257 | 258 | 可参考刘知远老师给出的清单 [如何不出国门走进NLP学术前沿（刘知远）](https://zhuanlan.zhihu.com/p/35380020) 259 | 260 | 主要有：中国中文信息学会（CIPS）中国计算机学会（CCF) 中国人工智能学会（CAAI）等 261 | 262 | 1. [KEQA 2019(南京大学)](http://ws.nju.edu.cn/conf/keqa2019/) 263 | 1. [Natural Language Question Answering over Knowledge Graph](http://ws.nju.edu.cn/conf/keqa2019/resources/%E9%82%B9%E7%A3%8A%20--%20Question%20Answering%20Over%20Knowledge%20Graph.pdf) 邹磊 (北京大学） 264 | 2. [Matching Questions and Answers in Dialogues from Online Forums](http://ws.nju.edu.cn/conf/keqa2019/resources/%E6%9C%B1%E5%85%B6%E7%AB%8B%20--%20Matching%20Questions%20and%20Answers%20in%20Dialogues%20from%20Online%20Forums-more.pdf) 朱其立 (上海交通大学） 265 | 3. [知识图谱中的关联搜索](http://ws.nju.edu.cn/conf/keqa2019/resources/%E7%A8%8B%E9%BE%9A%20--%20%E7%9F%A5%E8%AF%86%E5%9B%BE%E8%B0%B1%E4%B8%AD%E7%9A%84%E5%85%B3%E8%81%94%E6%90%9C%E7%B4%A2.pdf) 程龚 (南京大学） 266 | 4. [Understanding User Generated Information](http://ws.nju.edu.cn/conf/keqa2019/resources/%E9%92%B1%E9%93%81%E4%BA%91%20--%20Understanding%20User%20Generated%20Data.pdf) 钱铁云 (武汉大学） 267 | 5. [事理图谱的构建及应用](http://ws.nju.edu.cn/conf/keqa2019/resources/%E4%B8%81%E6%95%88%20--%20%E4%BA%8B%E7%90%86%E5%9B%BE%E8%B0%B1%E7%9A%84%E6%9E%84%E5%BB%BA%E5%8F%8A%E5%BA%94%E7%94%A8.pdf) 丁效 (哈尔滨工业大学） 268 | 6. [搜索引擎中的智能问答](http://ws.nju.edu.cn/conf/keqa2019/resources/%E5%BC%A0%E5%A5%87%20--%20%E6%90%9C%E7%B4%A2%E5%BC%95%E6%93%8E%E4%B8%AD%E7%9A%84%E6%99%BA%E8%83%BD%E9%97%AE%E7%AD%94.pdf) 张奇 (复旦大学） 269 | 2. CCF 学科前沿讲习班第 108 期 270 | 1. 大规模图神经网络与实践 杨红霞 （阿里巴巴） 271 | 2. 知识计算即服务：赋能企业知识化转型 袁晶（华为云） 272 | 3. [知识图谱融合方法](https://github.com/nju-websoft/KnowledgeGraphFusion) 胡伟(南京大学) 273 | 4. 生活领域知识图谱的构建及应用 张富峥（美团点评） 274 | 5. 知识图谱与众包数据库 李国良（清华大学） 275 | 6. 大规模知识图谱和自动化构建关键技术及应用 肖仰华（复旦大学） 276 | 3. [CIPS 前沿讲习班 2019](http://conference.cipsc.org.cn/ssatt2019/) 277 | 1. 面向自然语言处理的深度学习基础 邱锡鹏（复旦大学）颜航（复旦大学） 278 | 2. 开放域语义解析 韩先培（中国科学院软件研究所）陈波（中国科学院软件研究所） 279 | 3. 图神经网络在自然语言处理中的应用 张岳（西湖大学） 280 | 4. 基于深度学习的机器阅读理解 崔一鸣（科大讯飞） 281 | 5. 问答系统 唐都钰（微软亚洲研究院）段楠（微软亚洲研究院） 282 | 6. 任务型对话系统 车万翔（哈尔滨工业大学）车万翔（哈尔滨工业大学） 283 | 7. 人工智能在人机对话系统中的技术现状与挑战 严睿（北京大学） 284 | 4. [CIPS 前沿讲习班 2018](http://conference.cipsc.org.cn/ssatt2018/) 285 | 1. 语义表示学习 刘知远（清华大学） 286 | 2. 深度学习与词法句法语义分析 车万翔（哈尔滨工业大学） 287 | 3. 深度学习与机器翻译 张家俊（中国科学院自动化研究所）涂兆鹏（腾讯AI Lab） 288 | 4. 深度强化学习与GAN基础 俞扬（南京大学） 289 | 5. 信息检索中的深度强化学习新进展 徐君（中国科学院计算技术研究所）庞亮（中国科学院计算技术研究所） 290 | 6. 对话系统中的深度学习进展 李纪为（香侬科技） 291 | 7. 知识图谱中的深度学习新进展 William Wang （University California, Santa Barbara） 292 | 5. [CIPS 前沿讲习班 2017](http://www.cipsc.org.cn/att2017/) 293 | 1. 深度学习基础知识 邱锡鹏，复旦大学 294 | 2. 深度学习工具实战 龚经经，复旦大学 295 | 3. 深度学习与词法句法语义分析 车万翔，哈尔滨工业大学 296 | 4. 深度学习与知识获取 刘康，中科院自动化所 297 | 5. 深度学习与机器翻译 熊德意，苏州大学 298 | 6. 深度学习与自动问答 冯岩松，北京大学 299 | 7. 深度学习与社会计算 赵鑫，中国人民大学 300 | 8. 深度学习与信息检索 郭嘉丰，中科院计算所 301 | 6. 302 | --------------------------------------------------------------------------------