WinoGrande: an adversarial winograd schema challenge at scale
Open accessAbstract
Commonsense reasoning remains a major challenge in AI, and yet, recent progresses on benchmarks may seem to suggest otherwise. In particular, the recent neural language models have reported above 90% accuracy on the Winograd Schema Challenge (WSC), a commonsense benchmark originally designed to be unsolvable for statistical models that rely simply on word associations. This raises an important question---whether these models have truly acquired robust commonsense capabilities or they rely on spurious biases in the dataset 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, but adjusted to improve both the scale and the hardness of the dataset. The key steps of the dataset construction consist of (1) large-scale crowdsourcing, followed by (2) systematic bias reduction using a novel AFLITE algorithm that generalizes human-detectable word associations to machine-detectable embedding associations. Our experiments demonstrate that state-of-the-art models achieve considerably lower accuracy (59.4%-79.1%) on WINOGRANDE compared to humans (94%), confirming that the high performance on the original WSC was inflated by spurious biases in the dataset. Furthermore, we report new state-of-the-art results on five related benchmarks with emphasis on their dual implications. On the one hand, they demonstrate the effectiveness of WINOGRANDE when used as a resource for transfer learning. On the other hand, the high performance on all these benchmarks suggests the extent to which spurious biases are prevalent in all such datasets, which motivates further research on algorithmic bias reduction.
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?
Frontier models have achieved impressive performance in mathematics, coding, general knowledge, and commonsense reasoning.
“Frontier models have achieved impressive performance in mathematics, coding, general knowledge, and commonsense reasoning (Hendrycks et al., 2021a;b; Chen et al., 2021; Sakaguchi et al., 2021; Yue et al., 2024).”
From Does Spatial Cognition Emerge in Frontier Models? (2024) · cites this paper as Sakaguchi et al., 2021
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
I.—COMPUTING MACHINERY AND INTELLIGENCE
TURING, A. M. (1950) · Mind · doi:10.1093/mind/lix.236.433 · Cited by 10,290 · cited 1× · see citing works
“The Winograd Schema Challenge (WSC) (Levesque, Davis, and Morgenstern 2011), proposed as an alternative to the Turing Test (Turing 1950), has been used as a benchmark for evaluating commonsense reasoning.”
MethodUNIFIEDQA: 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
Commonsense reasoning and commonsense knowledge in artificial intelligence
Davis, E., & Marcus, G. (2015) · Communications of the ACM · doi:10.1145/2701413 · Cited by 434 · see citing works
Geva, M., Goldberg, Y., & Berant, J. (2019) · Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP) · doi:10.18653/v1/d19-1107 · Cited by 284 · see citing works
Flagged first, then by citations
Research works printed but not linked: 28
- Chen, X., Cardie, C. Multinomial adversarial networks for multi- domain text classification. NAACL (2018), 1226-1240.
- Clark, K., Manning, C.D. Deep reinforcement learning for mention-ranking coreference models. EMNLP (2016), 2256-2262.
- Devlin, J., Chang, M.-W., Lee, K., Toutanova, K. BERT: Pre-training of deep bidirectional transformers for language understanding. arXiv:1810.04805 (2018). · arXiv
- Durrett, G., Klein, D. Easy victories and uphill battles in coreference resolution. EMNLP (2013), 1971-1982.
- Elazar, Y., Goldberg, Y. Adversarial removal of demographic attributes from text data. EMNLP (2018), 11-21.
- Emami, A., Trischler, A., Suleman, K., Cheung, J.C.K. A generalized knowledge hunting framework for the winograd schema challenge. NAACL: SRW (2018), 25-31.
- Emami, A., Trichelair, P., Trischler, A., Suleman, K., Schulz, H., Cheung, J.C.K The KnowRef coreference corpus: Removing gender and number cues for difficult pronominal anaphora resolution. ACL (2019), 3952-3961.
- Gordon, A.S., Bejan, C.A., Sagae, K. Commonsense causal reasoning using millions of personal stories. AAAI (2011), 1180-1185.
- Gururangan, S., Swayamdipta, S., Levy, O., Schwartz, R., Bowman, S., Smith, N.A. Annotation artifacts in natural language inference data. NAACL (2018), 107-112.
- He, P., Liu, X., Chen, W., Gao, J. A hybrid neural network model for commonsense reasoning. arXiv:1907.11983 (2019). · arXiv
- Kocijan, V., Cretu, A.-M., Camburu, O.-M., Yordanov, Y., Lukasiewicz, T. A surprisingly robust trick for the winograd schema challenge. ACL (2019), 4837-4842.
- Le Bras, R., Swayamdipta, S., Bhagavatula, C., Zellers, R., Peters, M., Sabharwal, A., Choi, Y. Adversarial filters of dataset biases. ICML (2020).
- Levesque, H.J., Davis, E., Morgenstern, L. The winograd schema challenge. In AAAI Spring Symposium: Logical Formalizations of Commonsense Reasoning (2011).
- Lin, S.-C., Yang, J.-H., Nogueira, R., Tsai, M.-F., Wang, C.-J., Lin, J. Tttttackling winogrande schemas. arXiv preprint arXiv:2003.08380 (2020). · arXiv
- Liu, Q., Jiang, H., Ling, Z.-H., Zhu, X., Wei, S., Hu, Y. Commonsense knowledge enhanced embeddings for solving pronoun disambiguation problems in winograd schema challenge. arXiv:1611.04146 (2016). · arXiv
- Liu, Y., Ott, M., Goyal, N., Du, J., Joshi, M.S., Chen, D., Levy, O., Lewis, M., Zettlemoyer, L.S., Stoyanov, V. Roberta: A robustly optimized bert pretraining approach. ArXiv, abs/1907.11692 (2019). · arXiv
- Morgenstern, L., Davis, E., Ortiz, C. L. Planning, executing, and evaluating the winograd schema challenge. AI Magazine 37, 1 (2016), 50-54.
- Niven, T., Kao, H.-Y. Probing neural network comprehension of natural language arguments. ACL (2019), 4658-4664.
- Peng, H., Khashabi, D., Roth, D. Solving hard coreference problems. NAACL (2015), 809-819.
- Rahman, A., Ng, V. Resolving complex cases of definite pronouns: The winograd schema challenge. EMNLP-CoNLL (2012).
- Roemmele, M., Bejan, C.A., Gordon, A.S. Choice of plausible alternatives: An evaluation of commonsense causal reasoning. In AAAI Spring Symposium: Logical Formalizations of Commonsense Reasoning (2011).
- Rudinger, R., Naradowsky, J., Leonard, B., Van Durme, B. Gender bias in coreference resolution. NAACL (2018), 15-20.
- Stokes, P.D. Creativity from Constraints: The Psychology of Breakthrough. Springer Publishing Company, New York, NY, 2005.
- Trichelair, P., Emami, A., Cheung, J.C.K., Trischler, A., Suleman, K., Diaz, F. On the evaluation of commonsense reasoning in natural language understanding. arXiv:1811.01778 (2018). · arXiv
- Trinh, T.H., Le, Q.V. A simple method for commonsense reasoning. arXiv:1806.02847 (2018). · arXiv
- Tsuchiya, M. Performance impact caused by hidden bias of training data for recognizing textual entailment. LREC (2018), 1506-1511.
- Wang, A., Pruksachatkun, Y., Nangia, N., Singh, A., Michael, J., Hill, F., Levy, O., Bowman, S.R. Superglue: A stickier benchmark for general-purpose language understanding systems. arXiv:1905.00537 (2019). · arXiv
- Zellers, R., Bisk, Y., Schwartz, R., Choi, Y. Swag: A large-scale adversarial dataset for grounded commonsense inference. EMNLP (2018), 93-104.
Cited by
Sorted by citations
PaLM: Scaling Language Modeling with Pathways
Chowdhery, A., Narang, S., Devlin, J., Bosma, M., Mishra, G., Roberts, A., Barham, P., Chung, H. W., Sutton, C., Gehrmann, S., et al. (2022) · arXiv · doi:10.48550/arxiv.2204.02311 · Cited by 2,590 · see citing works
Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena
Zheng, L., Chiang, W.-L., Sheng, Y., Zhuang, S., Wu, Z., Zhuang, Y., Lin, Z., Li, Z., Li, D., Xing, E. P., et al. (2023) · arXiv · doi:10.48550/arxiv.2306.05685 · Cited by 1,888 · see citing works
A Comprehensive Overview of Large Language Models
Naveed, H., Khan, A. U., Qiu, S., Saqib, M., Anwar, S., Usman, M., Akhtar, N., Barnes, N., & Mian, A. (2025) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3744746 · Cited by 1,257 · see citing works
Mamba: Linear-Time Sequence Modeling with Selective State Spaces
Gu, A., & Dao, T. (2023) · arXiv · doi:10.48550/arxiv.2312.00752 · Cited by 1,579 · 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) · Advances in Neural Information Processing Systems 37 · doi:10.52202/079017-3417 · Cited by 286 · see citing works
Large language models and their impact in ophthalmology
Betzler, B. K., Chen, H., Cheng, C.-Y., Lee, C. S., Ning, G., Song, S. J., Lee, A. Y., Kawasaki, R., van Wijngaarden, P., Grzybowski, A., et al. (2023) · The Lancet Digital Health · doi:10.1016/s2589-7500(23)00201-7 · Cited by 161 · see citing works
Towards trustworthy LLMs: a review on debiasing and dehallucinating in large language models
Lin, Z., Guan, S., Zhang, W., Zhang, H., Li, Y., & Zhang, H. (2024) · Artificial Intelligence Review · doi:10.1007/s10462-024-10896-y · Cited by 141 · 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
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. (2023) · arXiv · doi:10.48550/arxiv.2308.08747 · Cited by 135 · see citing works
Wang, F., Zhang, Z., Zhang, X., Wu, Z., Mo, T., Lu, Q., Wang, W., Li, R., Xu, J., Tang, X., et al. (2025) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3768165 · Cited by 106 · see citing works
Testing AI on language comprehension tasks reveals insensitivity to underlying meaning
Dentella, V., Günther, F., Murphy, E., Marcus, G., & Leivada, E. (2024) · Scientific Reports · doi:10.1038/s41598-024-79531-8 · Cited by 54 · see citing works
DialFact: A Benchmark for Fact-Checking in Dialogue
Gupta, P., Wu, C.-S., Liu, W., & Xiong, C. (2022) · Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2022.acl-long.263 · Cited by 48 · see citing works
Scaling Data-Constrained Language Models
Muennighoff, N., Rush, A., Barak, B., Scao, T. L., Tazi, N., Piktus, A., Pyysalo, S., Wolf, T., & Raffel, C. (2023) · Advances in Neural Information Processing Systems 36 · doi:10.52202/075280-2191 · Cited by 47 · see citing works
MELTing Point: Mobile Evaluation of Language Transformers
Laskaridis, S., Katevas, K., Minto, L., & Haddadi, H. (2024) · Proceedings of the 30th Annual International Conference on Mobile Computing and Networking · doi:10.1145/3636534.3690668 · Cited by 37 · see citing works
An Open-Source Data Contamination Report for Large Language Models
Li, Y., Guo, Y., Guerin, F., & Lin, C. (2024) · Findings of the Association for Computational Linguistics: EMNLP 2024 · doi:10.18653/v1/2024.findings-emnlp.30 · Cited by 19 · see citing works
Titans: Learning to Memorize at Test Time
Behrouz, A., Zhong, P., & Mirrokni, V. (2025) · arXiv · doi:10.48550/arxiv.2501.00663 · Cited by 24 · see citing works
OLMo: Accelerating the Science of Language Models
Groeneveld, D., Beltagy, I., Walsh, P., Bhagia, A., Kinney, R., Tafjord, O., Jha, A. H., Ivison, H., Magnusson, I., Wang, Y., et al. (2024) · arXiv · doi:10.48550/arxiv.2402.00838 · Cited by 10 · see citing works
Effective Long-Context Scaling of Foundation Models
Xiong, W., Liu, J., Molybog, I., Zhang, H., Bhargava, P., Hou, R., Martin, L., Rungta, R., Sankararaman, K. A., Oguz, B., et al. (2023) · arXiv · doi:10.48550/arxiv.2309.16039 · Cited by 9 · see citing works
Observational Scaling Laws and the Predictability of Language Model Performance
Ruan, Y., Maddison, C. J., & Hashimoto, T. (2024) · arXiv · doi:10.48550/arxiv.2405.10938 · Cited by 3 · see citing works
Ivanova, A. A., Sathe, A., Lipkin, B., Kumar, U., Radkani, S., Clark, T. H., Kauf, C., Hu, J., Pramod, R. T., Grand, G., et al. (2024) · arXiv · doi:10.48550/arxiv.2405.09605 · Cited by 2
“We provide a flexible generative synthetic data pipeline, capable of producing many diverse datasets, each with unique specifications and statistics, while preserving metadata and decision traces.”
Nested Learning: The Illusion of Deep Learning Architectures
Behrouz, A., Razaviyayn, M., Zhong, P., & Mirrokni, V. (2025) · arXiv · doi:10.48550/arxiv.2512.24695 · Cited by 1 · see citing works
MergeMix: Optimizing Mid-Training Data Mixtures via Learnable Model Merging
Wang, J., Tian, C., Chen, K., Liu, Z., Mao, J., Zhao, W. X., Zhang, Z., & Zhou, J. (2026) · arXiv · doi:10.48550/arxiv.2601.17858
ShinkaEvolve: Towards Open-Ended And Sample-Efficient Program Evolution
Lange, R. T., Imajuku, Y., & Cetin, E. (2025) · arXiv · doi:10.48550/arxiv.2509.19349
ATLAS: Learning to Optimally Memorize the Context at Test Time
Behrouz, A., Li, Z., Kacham, P., Daliri, M., Deng, Y., Zhong, P., Razaviyayn, M., & Mirrokni, V. (2025) · arXiv · doi:10.48550/arxiv.2505.23735
Does Spatial Cognition Emerge in Frontier Models?
Ramakrishnan, S. K., Wijmans, E., Kraehenbuehl, P., & Koltun, V. (2024) · arXiv · doi:10.48550/arxiv.2410.06468
Behrouz, A., Razaviyayn, M., Zhong, P., & Mirrokni, V. (2025) · arXiv · doi:10.48550/arxiv.2504.13173
WSM: Decay-Free Learning Rate Schedule via Checkpoint Merging for LLM Pre-training
Tian, C., Wang, J., Zhao, Q., Chen, K., Liu, J., Liu, Z., Mao, J., Zhao, W. X., Zhang, Z., & Zhou, J. (2025) · arXiv · doi:10.48550/arxiv.2507.17634
Grattafiori, A., Dubey, A., Jauhri, A., Pandey, A., Kadian, A., Al-Dahle, A., Letman, A., Mathur, A., Schelten, A., Vaughan, A., et al. (2024) · arXiv · doi:10.48550/arxiv.2407.21783 · Cited by 505 · see citing works
STBLLM: Breaking the 1-Bit Barrier with Structured Binary LLMs
Dong, P., Li, L., Zhong, Y., Du, D., Fan, R., Chen, Y., Tang, Z., Wang, Q., Xue, W., Guo, Y., et al. (2024) · arXiv · doi:10.48550/arxiv.2408.01803 · Cited by 1 · see citing works
- Authors
- Sakaguchi, K., Bras, R. L., Bhagavatula, C., & Choi, Y.
- Affiliations
- Allen Institute for Artificial Intelligence (US)
- Allen Institute for AI, Seattle, WA
- University of Washington (US)
- Venue
- Communications of the ACM
- Published
- 2021-08-24
- Volume
- 64
- Issue
- 9
- Pages
- 99-106
- ISSN
- 0001-0782 (ISSN-L)
- 1557-7317
- OpenAlex
- W3194676777
- Cited by
- 356
- References
- 4
- Citations by year
- 2026: 105 · 2025: 135 · 2024: 73 · 2023: 32 · 2022: 9 · 2021: 2
- FWCI
- 10.25
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- Locations
- Communications of the ACM (publishedVersion, cc-by)
- Funders
- DARPA, MCS program (N66001-19-2-4031)
- NSF (IIS-1524371, IIS-1714566)
- DARPA, the CwC program (W911NF-15-1-0543)
- 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
- commonsense reasoning, Winograd Schema Challenge, word association, transfer learning, large language models, benchmark evaluation
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
