Elamur Research⌕
☰
home / papers / 2022 / CommonsenseQA 2.0: Exposing the Limits of AI through Gamification

CommonsenseQA 2.0: Exposing the Limits of AI through Gamification

Paper · 2022 · Alon Talmor, Ori Yoran, Ronan Le Bras et al.

Summary

Modern pre-trained language models reach human-level scores on many NLU benchmarks. Yet they remain fragile on adversarial inputs and still make basic commonsense mistakes, partly because static crowdsourced datasets contain exploitable artifacts. The authors turn data collection into a game. Players write yes/no questions or assertions meant to fool an AI opponent (RoBERTa-Large). They earn bonus points for including a topic concept drawn from ConceptNet and a relational phrase aimed at a particular reasoning skill. Other people validate each item, and authors lose points when validators disagree, which rewards unambiguous questions. The opponent model is retrained periodically so players cannot keep exploiting the same weakness. The resulting dataset is CommonsenseQA 2.0 (CSQA2), with 14,343 questions. Although the in-game model was modest in size, much larger systems struggle with the data: UNICORN-11B reaches 70.2%, while humans score 94.1%. The work suggests that gamified, model-in-the-loop collection can cheaply produce hard, diverse benchmarks while letting designers steer the content.

Key points

  • CSQA2 contains 14,343 yes/no items. Prompts came from the 1,875 top-ranked ConceptNet concepts and a hand-built list of 33 relational phrases covering causality, size, quantifiers, conditionals, and similar skills.
  • The in-game model started as RoBERTa-Large trained on 50K TWENTY QUESTIONS examples plus 73K assertions generated from ConceptNet triplets. It was retrained five times, at 1K, 2K, 5K, 10K and 20K collected questions.
  • Test accuracy: UNICORN-11B 70.2%, T5-11B 67.8%, few-shot GPT-3 52.9%, humans 94.1%.
  • On contrast sets (60 dev questions expanded to 276 perturbed variants), consistency was very low: UNICORN-11B 18.3, T5-11B 11.7, GPT-3 3.3.
  • Error analysis found that GPT-3 is weaker on causal reasoning and UNICORN on size comparisons. Adding Google search snippets helped long-tail factoid questions by up to 22.5 points.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓University of WashingtonSeattle · United States402 works · 212,762 citations🔬Allen Institute for Artificial IntelligenceSeattle · United States304 works · 106,138 citations🎓Tel Aviv UniversityTel Aviv · Israel89 works · 11,326 citations

 Fields, topics and concepts

📚 Shelves 5The field of science this work belongs to (OpenAlex fields).Computer Science · 18,614 worksArtificial Intelligence · 17,965 worksArts and Humanities · 17,450 worksLanguage and Linguistics · 17,296 worksComputer Vision and Pattern Recognition · 3,486 works

Cite

Cite as Alon Talmor, Ori Yoran, Ronan Le Bras, Chandra Bhagavatula, Yoav Goldberg, Yejin Choi and Jonathan Berant. 2022. CommonsenseQA 2.0: Exposing the Limits of AI through Gamification. arXiv:2201.05320.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.48550/arxiv.2201.05320 · arXiv 2201.05320

@misc{talmor2022commonsenseqa,
  title = {{CommonsenseQA 2.0: Exposing the Limits of AI through Gamification}},
  author = {Alon Talmor, Ori Yoran, Ronan {Le Bras}, Chandra Bhagavatula, Yoav Goldberg, Yejin Choi and Jonathan Berant},
  year = {2022},
  doi = {10.48550/arxiv.2201.05320},
  eprint = {2201.05320},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2201.05320}
}

References

47 references, 23 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Tom B. Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, Sandhini Agarwal, Ariel Herbert-Voss, Gretchen Krueger, Tom Henighan, Rewon Child, Aditya Ramesh, Daniel M. Ziegler, Jeffrey Wu, Clemens Winter,… — Language Models are Few-Shot Learners
  2. Alon Talmor and Jonathan Berant. MultiQA: An empirical investigation of generalization and transfer in reading comprehension. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 4911–4921, Florence, Italy, 2019. Association for Computational Linguistics.
  3. Adam Fisch, Alon Talmor, Robin Jia, Minjoon Seo, Eunsol Choi, and Danqi Chen. MRQA 2019 shared task: Evaluating generalization in reading comprehension. In Proceedings of the 2nd Workshop on Machine Reading for Question Answering, pages 1–13, Hong Kong, China, 2019. Association for Computational Lin…
  4. Robin Jia and Percy Liang. Adversarial examples for evaluating reading comprehension systems. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 2021–2031, Copenhagen, Denmark, 2017. Association for Computational Linguistics.
  5. Pranav Rajpurkar, Robin Jia, and Percy Liang. Know what you don’t know: Unanswerable questions for SQuAD. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 784–789, Melbourne, Australia, 2018. Association for Computational Ling…
  6. Xiaoyong Yuan, Pan He, Qile Zhu, and Xiaolin Li. Adversarial examples: Attacks and defenses for deep learning. IEEE transactions on neural networks and learning systems, 30(9):2805– 2824, 2019. find
  7. Marco Tulio Ribeiro, Sameer Singh, and Carlos Guestrin. Semantically equivalent adversarial rules for debugging NLP models. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 856–865, Melbourne, Australia, 2018. Association for C… find
  8. Matt Gardner, Yoav Artzi, Victoria Basmov, Jonathan Berant, Ben Bogin, Sihao Chen, Pradeep Dasigi, Dheeru Dua, Yanai Elazar, Ananth Gottumukkala, Nitish Gupta, Hannaneh Hajishirzi, Gabriel Ilharco, Daniel Khashabi, Kevin Lin, Jiangming Liu, Nelson F. Liu, Phoebe Mulcaire, Qiang Ning, Sameer Singh, N… — Evaluating Models' Local Decision Boundaries via Contrast Sets
  9. Divyansh Kaushik, Eduard H. Hovy, and Zachary Chase Lipton. Learning the difference that makes A difference with counterfactually-augmented data. In 8th International Conference on Learning Representations, ICLR 2020, Addis Ababa, Ethiopia, April 26-30, 2020. OpenReview.net, 2020. find
  10. Ludwig Schmidt, Shibani Santurkar, Dimitris Tsipras, Kunal Talwar, and Aleksander Madry. Adversarially robust generalization requires more data. In Samy Bengio, Hanna M. Wallach, Hugo Larochelle, Kristen Grauman, Nicolò Cesa-Bianchi, and Roman Garnett, editors, Advances in Neural Information Process… find
  11. Ronan Le Bras, Swabha Swayamdipta, Chandra Bhagavatula, Rowan Zellers, Matthew E. Peters, Ashish Sabharwal, and Yejin Choi. Adversarial filters of dataset biases. In Proceedings of the 37th International Conference on Machine Learning, ICML 2020, 13-18 July 2020, Virtual Event, volume 119 of Proceed… find
  12. Rowan Zellers, Yonatan Bisk, Roy Schwartz, and Yejin Choi. SWAG: A large-scale adversarial dataset for grounded commonsense inference. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 93–104, Brussels, Belgium, 2018. Association for Computational Lingu… find
  13. Rowan Zellers, Ari Holtzman, Yonatan Bisk, Ali Farhadi, and Yejin Choi. HellaSwag: Can a machine really finish your sentence? In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 4791–4800, Florence, Italy, 2019. Association for Computational Linguistics. find
  14. Maarten Sap, Hannah Rashkin, Derek Chen, Ronan Le Bras, and Yejin Choi. Social IQa: Commonsense reasoning about social interactions. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EM… find
  15. Keisuke Sakaguchi, Ronan Le Bras, Chandra Bhagavatula, and Yejin Choi. Winogrande: An adversarial winograd schema challenge at scale. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 34, pages 8732–8740, 2020. find
  16. Swabha Swayamdipta, Roy Schwartz, Nicholas Lourie, Yizhong Wang, Hannaneh Hajishirzi, Noah A. Smith, and Yejin Choi. Dataset cartography: Mapping and diagnosing datasets with training dynamics. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 9… find
  17. Lifu Tu, Garima Lalwani, Spandana Gella, and He He. An empirical study on robustness to spurious correlations using pre-trained language models. Transactions of the Association for Computational Linguistics, 8:621–633, 2020. find
  18. Danqi Chen, Jason Bolton, and Christopher D. Manning. A thorough examination of the CNN/Daily Mail reading comprehension task. In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 2358–2367, Berlin, Germany, 2016. Association for C…
  19. Sewon Min, Eric Wallace, Sameer Singh, Matt Gardner, Hannaneh Hajishirzi, and Luke Zettlemoyer. Compositional questions do not necessitate multi-hop reasoning. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 4249–4257, Florence, Italy, 2019. Associat…
  20. Karl Moritz Hermann, Tomás Kociský, Edward Grefenstette, Lasse Espeholt, Will Kay, Mustafa Suleyman, and Phil Blunsom. Teaching machines to read and comprehend. In Corinna Cortes, Neil D. Lawrence, Daniel D. Lee, Masashi Sugiyama, and Roman Garnett, editors, Advances in Neural Information Processing…
  21. Douwe Kiela, Max Bartolo, Yixin Nie, Divyansh Kaushik, Atticus Geiger, Zhengxuan Wu, Bertie Vidgen, Grusha Prasad, Amanpreet Singh, Pratik Ringshia, Zhiyi Ma, Tristan Thrush, Sebastian Riedel, Zeerak Waseem, Pontus Stenetorp, Robin Jia, Mohit Bansal, Christopher Potts, and Adina Williams. Dynabench:… find
  22. Max Bartolo, Alastair Roberts, Johannes Welbl, Sebastian Riedel, and Pontus Stenetorp. Beat the AI: Investigating adversarial human annotation for reading comprehension. Transactions of the Association for Computational Linguistics, 8:662–678, 2020.
  23. Mor Geva, Daniel Khashabi, Elad Segal, Tushar Khot, Dan Roth, and Jonathan Berant. Did aristotle use a laptop? a question answering benchmark with implicit reasoning strategies. arXiv preprint arXiv:2101.02235, 2021. find
  24. Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, and Veselin Stoyanov. Roberta: A robustly optimized bert pretraining approach. arXiv preprint arXiv:1907.11692, 2019.
  25. Nicholas Lourie, Ronan Le Bras, Chandra Bhagavatula, and Yejin Choi. Unicorn on rainbow: A universal commonsense reasoning model on a new multitask benchmark. AAAI, 2021. find
  26. Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, Wei Li, and Peter J. Liu. Exploring the limits of transfer learning with a unified text-to-text transformer. Journal of Machine Learning Research, 21(140):1–67, 2020.
  27. Robyn Speer, Joshua Chin, and Catherine Havasi. Conceptnet 5.5: An open multilingual graph of general knowledge. In Satinder P. Singh and Shaul Markovitch, editors, Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence, February 4-9, 2017, San Francisco, California, USA, pages 4… find
  28. Julian Eisenschlos, Bhuwan Dhingra, Jannis Bulian, Benjamin Börschinger, and Jordan Boyd-Graber. Fool me twice: Entailment from Wikipedia gamification. In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, p… find
  29. Alon Talmor, Jonathan Herzig, Nicholas Lourie, and Jonathan Berant. CommonsenseQA: A question answering challenge targeting commonsense knowledge. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume…
  30. Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, and Illia Polosukhin. Attention is all you need. In Isabelle Guyon, Ulrike von Luxburg, Samy Bengio, Hanna M. Wallach, Rob Fergus, S. V. N. Vishwanathan, and Roman Garnett, editors, Advances in Ne…
  31. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. BERT: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volum…
  32. Alon Talmor, Yanai Elazar, Yoav Goldberg, and Jonathan Berant. oLMpics-on what language model pre-training captures. Transactions of the Association for Computational Linguistics, 8:743–758, 2020. — oLMpics — On what Language Model Pre-training Captures
  33. A. Talmor and J. Berant. The web as knowledge-base for answering complex questions. In North American Association for Computational Linguistics (NAACL), 2018. find
  34. Dheeru Dua, Yizhong Wang, Pradeep Dasigi, Gabriel Stanovsky, Sameer Singh, and Matt Gardner. DROP: A reading comprehension benchmark requiring discrete reasoning over paragraphs. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Hum…
  35. Alon Talmor, Ori Yoran, Amnon Catav, Dan Lahav, Yizhong Wang, Akari Asai, Gabriel Ilharco, Hannaneh Hajishirzi, and Jonathan Berant. Multimodalqa: Complex question answering over text, tables and images. arXiv preprint arXiv:2104.06039, 2021. — [arXiv:2104.06039](https://arxiv.org/abs/2104.06039) find
  36. Alon Talmor, Oyvind Tafjord, Peter Clark, Yoav Goldberg, and Jonathan Berant. Leap-of-thought: Teaching pre-trained models to systematically reason over implicit knowledge. Advances in Neural Information Processing Systems, 33, 2020.
  37. Kevin Lin, Oyvind Tafjord, Peter Clark, and Matt Gardner. Reasoning over paragraph effects in situations. In Proceedings of the 2nd Workshop on Machine Reading for Question Answering, pages 58–62, Hong Kong, China, 2019. Association for Computational Linguistics.
  38. Pradeep Dasigi, Nelson F. Liu, Ana Marasović, Noah A. Smith, and Matt Gardner. Quoref: A reading comprehension dataset with questions requiring coreferential reasoning. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conferen…
  39. Joshua Attenberg, Panos Ipeirotis, and Foster Provost. Beat the machine: Challenging humans to find a predictive model’s “unknown unknowns”. Journal of Data and Information Quality (JDIQ), 6(1):1–17, 2015. find
  40. Allyson Ettinger, Sudha Rao, Hal Daumé III, and Emily M. Bender. Towards linguistically generalizable NLP systems: A workshop and shared task. In Proceedings of the First Workshop on Building Linguistically Generalizable NLP Systems, pages 1–10, Copenhagen, Denmark, 2017. Association for Computation… find
  41. Zhilin Yang, Peng Qi, Saizheng Zhang, Yoshua Bengio, William Cohen, Ruslan Salakhutdinov, and Christopher D. Manning. HotpotQA: A dataset for diverse, explainable multi-hop question answering. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 2369–2380,…
  42. Zhilin Yang, Saizheng Zhang, Jack Urbanek, Will Feng, Alexander H. Miller, Arthur Szlam, Douwe Kiela, and Jason Weston. Mastering the dungeon: Grounded language learning by mechanical turker descent. In 6th International Conference on Learning Representations, ICLR 2018, Vancouver, BC, Canada, April… find
  43. Michael Chen, Mike D’Arcy, Alisa Liu, Jared Fernandez, and Doug Downey. CODAH: An adversarially-authored question answering dataset for common sense. In Proceedings of the 3rd Workshop on Evaluating Vector Space Representations for NLP, pages 63–69, Minneapolis, USA, 2019. Association for Computatio…
  44. Yixin Nie, Adina Williams, Emily Dinan, Mohit Bansal, Jason Weston, and Douwe Kiela. Adversarial NLI: A new benchmark for natural language understanding. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 4885–4901, Online, 2020. Association for Computa…
  45. Panagiotis G. Ipeirotis and Evgeniy Gabrilovich. Quizz: targeted crowdsourcing with a billion (potential) users. In Chin-Wan Chung, Andrei Z. Broder, Kyuseok Shim, and Torsten Suel, editors, 23rd International World Wide Web Conference, WWW ’14, Seoul, Republic of Korea, April 7-11, 2014, pages 143–… find
  46. Luis Von Ahn and Laura Dabbish. Labeling images with a computer game. In Proceedings of the SIGCHI conference on Human factors in computing systems, pages 319–326, 2004. find
  47. Seth Cooper, Firas Khatib, Adrien Treuille, Janos Barbero, Jeehyung Lee, Michael Beenen, Andrew Leaver-Fay, David Baker, Zoran Popović, et al. Predicting protein structures with a multiplayer online game. Nature, 466(7307):756–760, 2010. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/talmor-2022-commonsenseqa-2-0-exposing-limits/