WinoGrande: An Adversarial Winograd Schema Challenge at Scale
Open accessAbstract
The Winograd Schema Challenge (WSC) (Levesque, Davis, and Morgenstern 2011), a benchmark for commonsense reasoning, is a set of 273 expert-crafted pronoun resolution problems originally designed to be unsolvable for statistical models that rely on selectional preferences or word associations. However, recent advances in neural language models have already reached around 90% accuracy on variants of WSC. This raises an important question whether these models have truly acquired robust commonsense capabilities or whether they rely on spurious biases in the datasets that lead to an overestimation of the true capabilities of machine commonsense.To investigate this question, we introduce WinoGrande, a large-scale dataset of 44k problems, inspired by the original WSC design, but adjusted to improve both the scale and the hardness of the dataset. The key steps of the dataset construction consist of (1) a carefully designed crowdsourcing procedure, followed by (2) systematic bias reduction using a novel AfLite algorithm that generalizes human-detectable word associations to machine-detectable embedding associations. The best state-of-the-art methods on WinoGrande achieve 59.4 – 79.1%, which are ∼15-35% (absolute) below human performance of 94.0%, depending on the amount of the training data allowed (2% – 100% respectively).Furthermore, we establish new state-of-the-art results on five related benchmarks — WSC (→ 90.1%), DPR (→ 93.1%), COPA(→ 90.6%), KnowRef (→ 85.6%), and Winogender (→ 97.1%). These results have dual implications: on one hand, they demonstrate the effectiveness of WinoGrande when used as a resource for transfer learning. On the other hand, they raise a concern that we are likely to be overestimating the true capabilities of machine commonsense across all these benchmarks. We emphasize the importance of algorithmic bias reduction in existing and future benchmarks to mitigate such overestimation.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The sampling approach biased by the selection of baseline models leads to results that are more interesting (harder for many models) than those obtained by using sampling methods directly tied to the properties of a particular model.
“While this approach is clearly biased by the selection of baseline models, in our experience, it leads to results that are more interesting (i.e., harder for many models) than those obtained by using sampling methods directly tied to the properties of a particular model (such as those obtained by active learning or adversarial approaches used to improve model training such as [3, 15]).”
From Shopping Queries Dataset: A Large-Scale ESCI Benchmark for Improving Product Search (2022) · cites this paper as [15]
The citation is attached to the example of an adversarial approach, not to the citing authors’ comparison with their own sampling method. Section 3 (p. 3) identifies AFLITE as an adversarial filtering method, but describes its purpose as reducing bias in a benchmark, not improving model training. The source also does not compare AFLITE with the citing authors’ sampling approach. (LLM, conf=0.83)
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1423 · Cited by 48,293 · see citing works
Flagged first, then by citations
Research works printed but not linked: 28
- Bansal, M., and Klein, D. 2012. Coreference semantics from web features. ACL .
- Chen, X., and Cardie, C. 2018. Multinomial adversarial networks for multi-domain text classification. NAACL .
- Clark, K., and Manning, C. D. 2016. Deep reinforcement learning for mention-ranking coreference models. EMNLP .
- Durrett, G., and Klein, D. 2013. Easy victories and uphill battles in coreference resolution. EMNLP .
- Elazar, Y., and Goldberg, Y. 2018. Adversarial removal of demographic attributes from text data. EMNLP .
- Emami, A.; Trischler, A.; Suleman, K.; and Cheung, J. C. K. 2018. A generalized knowledge hunting framework for the winograd schema challenge. NAACL: SRW .
- Emami, A.; Trichelair, P.; Trischler, A.; Suleman, K.; Schulz, H.; and Cheung, J. C. K. 2019. The KnowRef coreference corpus: Removing gender and number cues for difficult pronominal anaphora resolution. ACL .
- Geva, M.; Goldberg, Y.; and Berant, J. 2019. Are we modeling the task or the annotator? an investigation of annotator bias in natural language understanding datasets. arXiv:1908.07898 . · arXiv
- Gordon, A. S.; Bejan, C. A.; and Sagae, K. 2011. Commonsense causal reasoning using millions of personal stories. AAAI .
- Gururangan, S.; Swayamdipta, S.; Levy, O.; Schwartz, R.; Bowman, S.; and Smith, N. A. 2018. Annotation artifacts in natural language inference data. NAACL .
- He, P.; Liu, X.; Chen, W.; and Gao, J. 2019. A hybrid neural network model for commonsense reasoning. arXiv:1907.11983 . · arXiv
- Kobdani, H.; Schuetze, H.; Schiehlen, M.; and Kamp, H. 2011. Bootstrapping coreference resolution using word associations. ACL .
- Kocijan, V.; Cretu, A.-M.; Camburu, O.-M.; Yordanov, Y.; and Lukasiewicz, T. 2019. A surprisingly robust trick for the winograd schema challenge. ACL .
- Levesque, H. J.; Davis, E.; and Morgenstern, L. 2011. The winograd schema challenge. AAAI Spring Symposium: Logical Formalizations of Commonsense Reasoning .
- Liu, Q.; Jiang, H.; Ling, Z.-H.; Zhu, X.; Wei, S.; and Hu, Y. 2016. Commonsense knowledge enhanced embeddings for solving pronoun disambiguation problems in winograd schema challenge. arXiv:1611.04146 . · arXiv
- Liu, Y.; Ott, M.; Goyal, N.; Du, J.; Joshi, M. S.; Chen, D.; Levy, O.; Lewis, M.; Zettlemoyer, L. S.; and Stoyanov, V. 2019. Roberta: A robustly optimized bert pretraining approach. ArXiv abs/1907.11692. · arXiv
- Morgenstern, L.; Davis, E.; and Ortiz, C. L. 2016. Planning, executing, and evaluating the winograd schema challenge. AI Magazine 37(1):50-54.
- Niven, T., and Kao, H.-Y. 2019. Probing neural network comprehension of natural language arguments. ACL .
- Peng, H.; Khashabi, D.; and Roth, D. 2015. Solving hard coreference problems. NAACL .
- Ratinov, L., and Roth, D. 2012. Learning-based multi-sieve co-reference resolution with knowledge. EMNLP-CoNLL .
- Roemmele, M.; Bejan, C. A.; and Gordon, A. S. 2011. Choice of plausible alternatives: An evaluation of commonsense causal reasoning. AAAI Spring Symposium: Logical Formalizations of Commonsense Reasoning .
- Rudinger, R.; Naradowsky, J.; Leonard, B.; and Van Durme, B. 2018. Gender bias in coreference resolution. NAACL .
- Stokes, P. D. 2005. Creativity from constraints: The psychology of breakthrough . Springer Publishing Company.
- Trichelair, P.; Emami, A.; Cheung, J. C. K.; Trischler, A.; Suleman, K.; and Diaz, F. 2018. On the evaluation of common-sense reasoning in natural language understanding. arXiv:1811.01778 . · arXiv
- Trinh, T. H., and Le, Q. V. 2018. A simple method for commonsense reasoning. arXiv:1806.02847 . · arXiv
- Tsuchiya, M. 2018. Performance impact caused by hidden bias of training data for recognizing textual entailment. LREC .
- Wang, A.; Pruksachatkun, Y.; Nangia, N.; Singh, A.; Michael, J.; Hill, F.; Levy, O.; and Bowman, S. R. 2019. Superglue: A stickier benchmark for general-purpose language understanding systems. arXiv:1905.00537 . · arXiv
- Zellers, R.; Bisk, Y .; Schwartz, R.; and Choi, Y . 2018. Swag: A large-scale adversarial dataset for grounded commonsense inference. EMNLP .
Cited by
Sorted by citations
Exploiting Cloze-Questions for Few-Shot Text Classification and Natural Language Inference
Schick, T., & Schütze, H. (2021) · Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume · doi:10.18653/v1/2021.eacl-main.20 · Cited by 1,305 · see citing works
PIQA: Reasoning about Physical Commonsense in Natural Language
Bisk, Y., Zellers, R., Le bras, R., Gao, J., & Choi, Y. (2020) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v34i05.6239 · Cited by 37 · see citing works
Self-citationAdapterFusion: Non-Destructive Task Composition for Transfer Learning
Pfeiffer, J., Kamath, A., Rücklé, A., Cho, K., & Gurevych, I. (2021) · Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume · doi:10.18653/v1/2021.eacl-main.39 · Cited by 636 · see citing works
Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language models
Srivastava, A., Rastogi, A., Rao, A., Shoeb, A. A. M., Abid, A., Fisch, A., Brown, A. R., Santoro, A., Gupta, A., Garriga-Alonso, A., et al. (2022) · arXiv · doi:10.48550/arxiv.2206.04615 · Cited by 569 · see citing works
Self-citationUNIFIEDQA: Crossing Format Boundaries with a Single QA System
Khashabi, D., Min, S., Khot, T., Sabharwal, A., Tafjord, O., Clark, P., & Hajishirzi, H. (2020) · Findings of the Association for Computational Linguistics: EMNLP 2020 · doi:10.18653/v1/2020.findings-emnlp.171 · Cited by 514 · see citing works
CrowS-Pairs: A Challenge Dataset for Measuring Social Biases in Masked Language Models
Nangia, N., Vania, C., Bhalerao, R., & Bowman, S. R. (2020) · Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2020.emnlp-main.154 · Cited by 446 · see citing works
Cross-Task Generalization via Natural Language Crowdsourcing Instructions
Mishra, S., Khashabi, D., Baral, C., & Hajishirzi, H. (2022) · Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2022.acl-long.244 · Cited by 269 · see citing works
Super-NaturalInstructions: Generalization via Declarative Instructions on 1600+ NLP Tasks
Wang, Y., Mishra, S., Alipoormolabashi, P., Kordi, Y., Mirzaei, A., Naik, A., Ashok, A., Dhanasekaran, A. S., Arunkumar, A., Stap, D., et al. (2022) · Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2022.emnlp-main.340 · Cited by 247 · see citing works
BOLD: Dataset and Metrics for Measuring Biases in Open-Ended Language Generation
Dhamala, J., Sun, T., Kumar, V., Krishna, S., Pruksachatkun, Y., Chang, K.-W., & Gupta, R. (2021) · Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency · doi:10.1145/3442188.3445924 · Cited by 244 · see citing works
Generative AI in Higher Education: The ChatGPT Effect
Chan, C. K. Y., & Colloton, T. (2024) · Routledge · doi:10.4324/9781003459026 · Cited by 203 · see citing works
Jiang, Z., Araki, J., Ding, H., & Neubig, G. (2021) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00407 · Cited by 177 · see citing works
MetaICL: Learning to Learn In Context
Min, S., Lewis, M., Zettlemoyer, L., & Hajishirzi, H. (2022) · Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies · doi:10.18653/v1/2022.naacl-main.201 · Cited by 169 · see citing works
Symbolic Discovery of Optimization Algorithms
Chen, X., Liang, C., Huang, D., Real, E., Wang, K., Liu, Y., Pham, H., Dong, X., Luong, T., Hsieh, C.-J., et al. (2023) · arXiv · doi:10.48550/arxiv.2302.06675 · Cited by 303 · see citing works
An Empirical Study of Catastrophic Forgetting in Large Language Models During Continual Fine-Tuning
Luo, Y., Yang, Z., Meng, F., Li, Y., Zhou, J., & Zhang, Y. (2025) · IEEE Transactions on Audio, Speech and Language Processing · doi:10.1109/taslpro.2025.3606231 · Cited by 136 · see citing works
Word meaning in minds and machines.
Lake, B. M., & Murphy, G. L. (2023) · Psychological Review · doi:10.1037/rev0000297 · Cited by 131 · see citing works
Abstraction and analogy‐making in artificial intelligence
Mitchell, M. (2021) · Annals of the New York Academy of Sciences · doi:10.1111/nyas.14619 · Cited by 125 · see citing works
Neural Theory-of-Mind? On the Limits of Social Intelligence in Large LMs
Sap, M., Le Bras, R., Fried, D., & Choi, Y. (2022) · Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2022.emnlp-main.248 · Cited by 118 · see citing works
Self-citationDeepNet: Scaling Transformers to 1,000 Layers
Wang, H., Ma, S., Dong, L., Huang, S., Zhang, D., & Wei, F. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3386927 · Cited by 113 · see citing works
Few-Shot Text Generation with Natural Language Instructions
Schick, T., & Schütze, H. (2021) · Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2021.emnlp-main.32 · Cited by 107 · see citing works
xLSTM: Extended Long Short-Term Memory
Beck, M., Pöppel, K., Spanring, M., Auer, A., Prudnikova, O., Kopp, M., Klambauer, G., Brandstetter, J., & Hochreiter, S. (2024) · arXiv · doi:10.48550/arxiv.2405.04517 · Cited by 299 · see citing works
Large language models for life cycle assessments: Opportunities, challenges, and risks
Preuss, N., Alshehri, A. S., & You, F. (2024) · Journal of Cleaner Production · doi:10.1016/j.jclepro.2024.142824 · Cited by 71 · see citing works
Data and its (dis)contents: A survey of dataset development and use in machine learning research
Paullada, A., Raji, I. D., Bender, E. M., Denton, E., & Hanna, A. (2021) · Patterns · doi:10.1016/j.patter.2021.100336 · Cited by 548 · see citing works
What Language Model to Train if You Have One Million GPU Hours?
Le Scao, T., Wang, T., Hesslow, D., Bekman, S., Bari, M. S., Biderman, S., Elsahar, H., Muennighoff, N., Phang, J., Press, O., et al. (2022) · Findings of the Association for Computational Linguistics: EMNLP 2022 · doi:10.18653/v1/2022.findings-emnlp.54 · Cited by 33 · see citing works
UPRISE: Universal Prompt Retrieval for Improving Zero-Shot Evaluation
Cheng, D., Huang, S., Bi, J., Zhan, Y., Liu, J., Wang, Y., Sun, H., Wei, F., Deng, W., & Zhang, Q. (2023) · Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2023.emnlp-main.758 · Cited by 32 · see citing works
Digital forgetting in large language models: a survey of unlearning methods
Blanco-Justicia, A., Jebreel, N., Manzanares-Salor, B., Sánchez, D., Domingo-Ferrer, J., Collell, G., & Eeik Tan, K. (2025) · Artificial Intelligence Review · doi:10.1007/s10462-024-11078-6 · Cited by 31 · see citing works
Understanding Commonsense Inference Aptitude of Deep Contextual Representations
Da, J., & Kasai, J. (2019) · Proceedings of the First Workshop on Commonsense Inference in Natural Language Processing · doi:10.18653/v1/d19-6001 · Cited by 22 · see citing works
FANToM: A Benchmark for Stress-testing Machine Theory of Mind in Interactions
Kim, H., Sclar, M., Zhou, X., Bras, R., Kim, G., Choi, Y., & Sap, M. (2023) · Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2023.emnlp-main.890 · Cited by 27 · see citing works
Self-citationDon't Make Your LLM an Evaluation Benchmark Cheater
Zhou, K., Zhu, Y., Chen, Z., Chen, W., Zhao, W. X., Chen, X., Lin, Y., Wen, J.-R., & Han, J. (2023) · arXiv · doi:10.48550/arxiv.2311.01964 · Cited by 25 · see citing works
Nezhurina, M., Cipolina-Kun, L., Cherti, M., & Jitsev, J. (2024) · arXiv · doi:10.48550/arxiv.2406.02061 · Cited by 21 · see citing works
Comparing Test Sets with Item Response Theory
Vania, C., Htut, P. M., Huang, W., Mungra, D., Pang, R. Y., Phang, J., Liu, H., Cho, K., & Bowman, S. R. (2021) · Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers) · doi:10.18653/v1/2021.acl-long.92 · Cited by 18 · see citing works
Shopping Queries Dataset: A Large-Scale ESCI Benchmark for Improving Product Search
Reddy, C. K., Màrquez, L., Valero, F., Rao, N., Zaragoza, H., Bandyopadhyay, S., Biswas, A., Xing, A., & Subbian, K. (2022) · arXiv · doi:10.48550/arxiv.2206.06588 · Cited by 17 · see citing works
tinyBenchmarks: evaluating LLMs with fewer examples
Polo, F. M., Weber, L., Choshen, L., Sun, Y., Xu, G., & Yurochkin, M. (2024) · arXiv · doi:10.48550/arxiv.2402.14992 · Cited by 8 · see citing works
Palmeira Ferraz, T., Mehta, K., Lin, Y.-H., Chang, H.-S., Oraby, S., Liu, S., Subramanian, V., Chung, T., Bansal, M., & Peng, N. (2024) · Findings of the Association for Computational Linguistics: EMNLP 2024 · doi:10.18653/v1/2024.findings-emnlp.458 · Cited by 3 · see citing works
Gradient Ascent Post-training Enhances Language Model Generalization
Yoon, D., Jang, J., Kim, S., & Seo, M. (2023) · Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers) · doi:10.18653/v1/2023.acl-short.74 · Cited by 3 · see citing works
OLMo, T., Walsh, P., Soldaini, L., Groeneveld, D., Lo, K., Arora, S., Bhagia, A., Gu, Y., Huang, S., Jordan, M., et al. (2025) · arXiv · doi:10.48550/arxiv.2501.00656 · Cited by 2 · see citing works
ARB-LLM: Alternating Refined Binarizations for Large Language Models
Li, Z., Yan, X., Zhang, T., Qin, H., Xie, D., Tian, J., shi, Z., Kong, L., Zhang, Y., & Yang, X. (2024) · arXiv · doi:10.48550/arxiv.2410.03129 · Cited by 1 · see citing works
MesaNet: Sequence Modeling by Locally Optimal Test-Time Training
von Oswald, J., Scherrer, N., Kobayashi, S., Versari, L., Yang, S., Mittal, S., Schlegel, M., Maile, K., Schimpf, Y., Sieberling, O., et al. (2025) · arXiv
On Effectiveness and Efficiency of Agentic Tool-calling and RL Training
Liu, T., Qian, C., Cief, M., He, Y., Dan, D., Aletras, N., & Kazai, G. (2026) · arXiv
Gated DeltaNet-2: Decoupling Erase and Write in Linear Attention
Hatamizadeh, A., Choi, Y., & Kautz, J. (2026) · arXiv
Self-citationSignal and Noise: A Framework for Reducing Uncertainty in Language Model Evaluation
Heineman, D., Hofmann, V., Magnusson, I., Gu, Y., Smith, N. A., Hajishirzi, H., Lo, K., & Dodge, J. (2025) · arXiv · doi:10.48550/arxiv.2508.13144
LittleBit: Ultra Low-Bit Quantization via Latent Factorization
Lee, B., Kim, D., You, Y., & Kim, Y. (2025) · arXiv · doi:10.48550/arxiv.2506.13771
- Authors
- Sakaguchi, K., Le Bras, R., Bhagavatula, C., & Choi, Y.
- Affiliations
- Allen Institute for Artificial Intelligence (US)
- University of Washington (US)
- Venue
- Proceedings of the AAAI Conference on Artificial Intelligence
- Published
- 2020-04-03
- Volume
- 34
- Issue
- 05
- Pages
- 8732-8740
- ISSN
- 2159-5399 (ISSN-L)
- 2374-3468
- OpenAlex
- W2996908057
- Cited by
- 579
- References
- 1
- Citations by year
- 2026: 80 · 2025: 116 · 2024: 50 · 2023: 92 · 2022: 67 · 2021: 100 · 2020: 66 · 2019: 8
- FWCI
- 30.42
- Type
- conference paper
- Language
- en
- Open access
- diamond
- doi.org
- Licence
- www.aaai.org/ (unspecified)
- Locations
- Proceedings of the AAAI Conference on Artificial Intelligence (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- AAAI Publications (The Association for the Advancement of Artificial Intelligence (AAAI)) (submittedVersion)
- arXiv (Cornell University)
- Funders
- National Science Foundation (1524371, 1714566)
- Defense Advanced Research Projects Agency (N66001, N66001-19-2-4031, N66001-19)
- Army Research Office (W911NF-15-1-, W911NF, W911NF-15)
- Naval Information Warfare Center Pacific
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Multimodal Machine Learning Applications
- Keywords
- Winograd Schema Challenge, commonsense reasoning, pronoun resolution, dataset construction, bias mitigation, word association, spurious correlations, transfer learning, COPA, DPR, large-scale datasets
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
