交大ADAPT实验室 | 常识知识论文列表
近年来，常识知识(commonsense knowledge)越发成为NLP领域，乃至多模态，跨学科的研究热点。尽管以BERT为代表的的预训练语言模型已经在实体识别、机器翻译、情感分析等任务上取得了亮眼的表现，它们在面对一些人类可以使用常识轻易解决的问题时仍然表现不佳，面对对抗样本时也极为脆弱。似乎这些模型知识只是学到了一些浅层线索和语义，而与人类的知识体系并不相同。因此，将常识知识融入机器学习中就成为了一个充满前景的解决方案。不过，如何抽取常识知识，如何利用常识知识，乃至如何定义“常识本身”依然是有待研究的难题。
交大ADAPT实验室近年来正将研究兴趣专注于这一领域。在此过程中，我们也将自己的阅读加以筛选、归类和统计，整合成了一份论文列表，在GitHub上开源。
https://github.com/adapt-sjtu/commonsense-papers
此paper-list精选近年来有代表性的论文、survey、tutorial，以及这一领域经典著作、资源。并将随着领域进展不断更新，也欢迎大家的意见和贡献。
以下是这一份paper-list在本文写作时刻的一张快照，如果想关注这份列表的实时更新，欢迎star和watch我们的Github repo.
commonsense-papers
Must-read papers on commonsense knowledge and others resources and tutorials
We aim to select the most representative and innovative papers in the research field of commonsense knowledge, and provide taxonomy/classification as well as statistics of these papers to give a quick overview of the field and help focused reading.
Contributed by ADAPTers (major efforts by Zhiling Zhang(@blmoistawinde), Siyu Ren, Hongru Huang, Zelin Zhou, Yanzhu Guo)
Our list may not be complete. We will keep adding papers and improving it. Contributions are welcomed!
Table of Contents
- commonsense-papers
- Statistics
- Tutorial and Survey
- Resources and Evaluation
- Commonsense Knowledge Bases/Models
- Related Knowledge Bases
- Datasets and Benchmarks
- Evaluation and Probing
- Knowledge Mining (Knowledge Base Completion)
- Applications
- natural language inference (NLI) and commonsense reasoning (CSR)
- Machine Reading Comprehension (MRC)
- Generation
- Question Answering (QA)
- Vision, Robotics, Multimodal, Grounding and Speech
Statistics
Total papers in this repo: 52 .
Research Keywords
Non-stopping words in title, indicating the hot topics in this field.
| count | |
|---|---|
| reasoning | 12 |
| graph | 5 |
| model | 5 |
| question | 5 |
| multi | 3 |
| pre | 3 |
| machine | 3 |
| reason | 3 |
| answering | 3 |
| story | 3 |
Researchers
Most active researchers in this field
| count | |
|---|---|
| Yejin Choi | 13 |
| Antoine Bosselut | 7 |
| Chandra Bhagavatula | 7 |
| Ronan Le Bras | 5 |
| Bill Yuchen Lin | 5 |
| Dan Roth | 4 |
| Maarten Sap | 4 |
| Hannah Rashkin | 4 |
| Vered Shwartz | 3 |
| Xiang Ren | 3 |
Venues
Just an estimation. May not be precise as arxiv papers may appear in other venues in the future.
| count | |
|---|---|
| ACL | 10 |
| EMNLP | 10 |
| arxiv | 7 |
| AAAI | 5 |
| CVPR | 2 |
Tutorial and Survey
Recent Advances in Natural Language Inference: A Survey of Benchmarks, Resources, and Approaches arxiv 2019 paper
Shane Storks, Qiaozi Gao, Joyce Y. Chai
T6: Commonsense Reasoning for Natural Language Processing. ACL 2020. slides and video
Antoine Bosselut, Dan Roth, Maarten Sap, Vered Shwartz, Yejin Choi
back to table of contents
Resources and Evaluation
Commonsense Knowledge Bases/Models
CYC: Toward programs with common sense data(OpenCyc) homepage
Commonsense causal reasoning between short texts (CausalNet) KR 2016 paper homepage code
Zhiyi Luo, Yuchen Sha, Kenny Q. Zhu, Seung-won Hwang, Zhongyuan Wang
ConceptNet 5.5: an open multilingual graph of general knowledge AAAI 2017. paper homepage repo embedding(ConceptNet Numberbatch) repo
Robyn Speer, Joshua Chin, Catherine Havasi
ATOMIC: An Atlas of Machine Commonsense for If-Then Reasoning AAAI 2019. paper homepage
Maarten Sap, Ronan LeBras, Emily Allaway, Chandra Bhagavatula, Nicholas Lourie, Hannah Rashkin, Brendan Roof, Noah A. Smith, Yejin Choi
COMET: Commonsense Transformers for Automatic Knowledge Graph Construction ACL 2019 paper homepage code
Antoine Bosselut, Hannah Rashkin, Maarten Sap, Chaitanya Malaviya, Asli Celikyilmaz, Yejin Choi
ASER: A Large-scale Eventuality Knowledge Graph. WWW 2020 paper homepage code
Hongming Zhang, Xin Liu, Haojie Pan, Yangqiu Song, Cane Wing-Ki Leung
Commonsense Properties from Query Logs and Question Answering Forums (quasimodo) CIKM 2019 paper homepage code
Julien Romero, Simon Razniewski, Koninika Pal, Jeff Z. Pan, Archit Sakhadeo, Gerhard Weikum
TransOMCS: From Linguistic Graphs to Commonsense Knowledge IJCAI 2020 paper homepage code
Hongming Zhang, Daniel Khashabi, Yangqiu Song, Dan Roth
Paragraph-Level Commonsense Transformers with Recurrent Memory (PARA-COMeT) arxiv 2020 paper
Saadia Gabriel, Chandra Bhagavatula, Vered Shwartz, Ronan Le Bras, Maxwell Forbes, Yejin Choi
COMET-ATOMIC 2020: On Symbolic and Neural Commonsense Knowledge Graphs arxiv 2020 paper
Jena D. Hwang, Chandra Bhagavatula, Ronan Le Bras, Jeff Da, Keisuke Sakaguchi, Antoine Bosselut, Yejin Choi
CommonGen: A Constrained Text Generation Challenge for Generative Commonsense Reasoning EMNLP 2020 Findings paper homepage
Bill Yuchen Lin, Wangchunshu Zhou, Ming Shen, Pei Zhou, Chandra Bhagavatula, Yejin Choi, Xiang Ren
Related Knowledge Bases
WordNet: A Lexical Database for English Communications of the ACM Vol. 38, No. 11: 39-41. 1995. [homepage] (https://wordnet.princeton.edu/)
George A. Miller
Toward an Architecture for Never-Ending Language Learning (NELL). AAAI 2010 paper homepage
Andrew Carlson, Justin Betteridge, Bryan Kisiel, Burr Settles, Estevam R. Hruschka Jr., and Tom M. Mitchell
DBpedia – A Large-scale, Multilingual Knowledge Base Extracted from Wikipedia. Semantic Web 2012 paper homepage
Jens Lehmann, Robert Isele, Max Jakob, Anja Jentzsch, Dimitris Kontokostas, Pablo N. Mendes, Sebastian Hellmann, Mohamed Morsey, Patrick van Kleef, Sören Auer, Christian Bizer
Datasets and Benchmarks
A Corpus and Cloze Evaluation for Deeper Understanding of Commonsense Stories (Story Cloze Test) NAACL 2016 paper homepage
Nasrin Mostafazadeh, Nathanael Chambers, Xiaodong He, Devi Parikh, Dhruv Batra, Lucy Vanderwende, Pushmeet Kohli, James Allen
Modeling Naive Psychology of Characters in Simple Commonsense Stories ACL 2018 paper homepage
Hannah Rashkin, Antoine Bosselut, Maarten Sap, Kevin Knight, Yejin Choi
SemEval-2018 Task 11: Machine Comprehension Using Commonsense Knowledge SemEval 2018 paper
Simon Ostermann, Michael Roth, Ashutosh Modi, Stefan Thater, Manfred Pinkal
Reasoning about Actions and State Changes by Injecting Commonsense Knowledge (ProPara) EMNLP 2018 paper code
Niket Tandon, Bhavana Dalvi Mishra, Joel Grus, Wen-tau Yih, Antoine Bosselut, Peter Clark
CommonSenseqa: A question answering challenge targeting commonsense knowledge NAACL 2019 paper homepage code
Alon Talmor, Jonathan Herzig, Nicholas Lourie, Jonathan Berant
From Recognition to Cognition: Visual Commonsense Reasoning CVPR 2019 paper code homepage
Rowan Zellers, Yonatan Bisk, Ali Farhadi, Yejin Choi
oLMpics – On what Language Model Pre-training Captures arxiv 2019 paper code homepage
Alon Talmor, Yanai Elazar, Yoav Goldberg, Jonathan Berant
ABDUCTIVE COMMONSENSE REASONING (ART) ICLR 2020 paper homepage
Chandra Bhagavatula, Ronan Le Bras, Chaitanya Malaviya, Keisuke Sakaguchi, Ari Holtzman, Hannah Rashkin, Doug Downey, Scott Wen-tau Yih, Yejin Choi
PIQA: Reasoning about Physical Commonsense in Natural Language AAAI 2020 paper homepage
Yonatan Bisk, Rowan Zellers, Ronan Le Bras, Jianfeng Gao, Yejin Choi
Reasoning about Goals, Steps, and Temporal Ordering with WikiHow EMNLP 2020 paper code
Qing Lyu, Li Zhang, Chris Callison-Burch
Birds have four legs?! NumerSense: Probing Numerical Commonsense Knowledge of Pre-trained Language Models EMNLP 2020 paper homepage
Bill Yuchen Lin, Seyeon Lee, Rahul Khanna and Xiang Ren
back to table of contents
Evaluation and Probing
Reporting Bias and Knowledge Acquisition AKBC 2013 paper
Jonathan Gordon, Benjamin Van Durme
Joint Reasoning for Multi-Faceted Commonsense Knowledge AKBC 2020 paper homepage
Yohan Chalier, Simon Razniewski, Gerhard Weikum
Does BERT Solve Commonsense Task via Commonsense Knowledge? arxiv 2020 paper
Leyang Cui, Sijie Cheng, Yu Wu, Yue Zhang
back to table of contents
Knowledge Mining (Knowledge Base Completion)
Automatic Extraction of Commonsense LocatedNear Knowledge ACL 2018 paper code
Frank F. Xu, Bill Yuchen Lin, Kenny Q. Zhu
Commonsense Knowledge Mining from Pretrained Models EMNLP 2019 paper
Joe Davison, Joshua Feldman, Alexander Rush
Language Models as Knowledge Bases? EMNLP 2019 paper code
Fabio Petroni, Tim Rocktäschel, Patrick Lewis, Anton Bakhtin, Yuxiang Wu, Alexander H. Miller, Sebastian Riedel
Commonsense Knowledge Base Completion with Structural and Semantic Context AAAI 2020 paper code
Chaitanya Malaviya, Chandra Bhagavatula, Antoine Bosselut, Yejin Choi
Temporal Common Sense Acquisition with Minimal Supervision ACL 2020 paper
Ben Zhou, Qiang Ning, Daniel Khashabi, Dan Roth
back to table of contents
Applications
natural language inference (NLI) and commonsense reasoning (CSR)
Neural natural language inference models enhanced with external knowledge ACL 2018 paper code
- resource: WordNet
Qian Chen, Xiaodan Zhu, Zhen-Hua Ling, Diana Inkpen, Si Wei
Incorporating Structured Commonsense Knowledge in Story Completion AAAI 2019 paper
- resource: ConceptNet Numberbatch
Jiaao Chen, Jianshu Chen, Zhou Yu
Contrastive Self-Supervised Learning for Commonsense Reasoning ACL 2020 paper code
Tassilo Klein, Moin Nabi
Do Language Embeddings Capture Scales? EMNLP Findings 2020 paper code
Xikun Zhang, Deepak Ramachandran, Ian Tenney, Yanai Elazar, Dan Roth
Differentiable Open-Ended Commonsense Reasoning arxiv 2020 paper
Bill Yuchen Lin, Haitian Sun, Bhuwan Dhingra, Manzil Zaheer, Xiang Ren, William W. Cohen
Machine Reading Comprehension (MRC)
Enhancing Pre-Trained Language Representations with Rich Knowledge for Machine Reading Comprehension ACL 2019 paper code
- resource: WordNet, NELL
An Yang, Quan Wang, Jing Liu, Kai Liu, Yajuan Lyu, Hua Wu, Qiaoqiao She, Sujian Li
Generation
Diverse and Informative Dialogue Generation with Context-Specific Commonsense Knowledge Awareness ACL 2020 paper code
- resource: ConceptNet
Sixing Wu, Ying Li, Dawei Zhang, Yang Zhou, Zhonghai Wu
Language Generation with Multi-Hop Reasoning on Commonsense Knowledge Graph EMNLP 2020 paper
- resource: ConceptNet
Haozhe Ji, Pei Ke, Shaohan Huang, Furu Wei, Xiaoyan Zhu, Minlie Huang
Question Answering (QA)
Dynamic Knowledge Graph Construction for Zero-shot Commonsense Question Answering arxiv 2019 paper
- resource: COMET
Antoine Bosselut, Yejin Choi
Adversarial Training for Commonsense Inference ACL 2020 RepL4NLP workshop paper
Lis Pereira, Xiaodong Liu, Fei Cheng, Masayuki Asahara, Ichiro Kobayashi
Unsupervised Commonsense Question Answering with Self-Talk EMNLP 2020 paper code
- resource: ConceptNet, COMET, Google N-grams
Vered Shwartz, Peter West, Ronan Le Bras, Chandra Bhagavatula, Yejin Choi
Scalable Multi-Hop Relational Reasoning for Knowledge-Aware Question Answering EMNLP 2020 paper code
Yanlin Feng, Xinyue Chen, Bill Yuchen Lin, Peifeng Wang, Jun Yan, Xiang Ren
Vision, Robotics, Multimodal, Grounding and Speech
Enabling Robots to Understand Incomplete Natural Language Instructions Using Commonsense Reasoning ICRA 2020 paper
Haonan Chen; Hao Tan; Alan Kuntz; Mohit Bansal; Ron Alterovitz
Visual Commonsense R-CNN CVPR 2020 paper code
Tan Wang, Jianqiang Huang, Hanwang Zhang, Qianru Sun
back to table of contents