Large Language Models are Better Reasoners with Self-Verification
Open accessAbstract
Recently, with the chain of thought (CoT) prompting, large language models (LLMs), e.g., GPT-3, have shown strong reasoning ability in several natural language processing tasks such as arithmetic, commonsense, and logical reasoning.However, LLMs with CoT require multi-step prompting and multi-token prediction, which is highly sensitive to individual mistakes and vulnerable to error accumulation.The above issues make the LLMs need the ability to verify the answers.In fact, after inferring conclusions in some thinking decision tasks, people often check them by re-verifying steps to avoid some mistakes.In this paper, we propose and prove that LLMs also have similar self-verification abilities.We take the conclusion obtained by CoT as one of the conditions for solving the original problem.By performing a backward verification of the answers that LLM deduced for itself, we can obtain interpretable answer validation scores to select the candidate answer with the highest score.Experimental results demonstrate that the proposed method can improve the reasoning performance on various arithmetic, commonsense, and logical reasoning datasets.Our code is publicly available at: https://github.com/WENGSYX/ Self-Verification.
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 belief that LLMs can self-critique and improve their own solutions in an iterative fashion rests on the assumption that verification of correctness should be easier than generation.
“This belief seemingly rests on the assumption that verification of correctness should be easier than generation-a rather classical argument from computational complexity”
From On the Self-Verification Limitations of Large Language Models on Reasoning and Planning Tasks (2024) · cites this paper as [50]
The paper’s analysis of verification accuracy reports that selecting a correct answer from candidates outperformed standalone chain-of-thought reasoning on six datasets. That supports a limited comparison between verification and generation, but the paper does not state that verification is generally easier than generation or ground its method in a computational-complexity argument. Its method reranks candidate answers rather than iteratively revising a solution. (LLM, conf=0.78)
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
Training language models to follow instructions with human feedback
Ouyang, L., Wu, J., Jiang, X., Almeida, D., Wainwright, C. L., Mishkin, P., Zhang, C., Agarwal, S., Slama, K., Ray, A., et al. (2022) · arXiv · doi:10.48550/arxiv.2203.02155 · Cited by 4,208 · 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-citationProgram Induction by Rationale Generation: Learning to Solve and Explain Algebraic Word Problems
Ling, W., Yogatama, D., Dyer, C., & Blunsom, P. (2017) · Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/p17-1015 · Cited by 325 · see citing works
Solving General Arithmetic Word Problems
Roy, S., & Roth, D. (2016) · arXiv · doi:10.48550/arxiv.1608.01413 · Cited by 275 · cited 4×
“Table 7 – Continued from previous page Dataset Question Self-Verification MultiArith Q: Luke was organizing his baseball cards in a binder with 3 on each page.”
InfluentialAutomatic Chain of Thought Prompting in Large Language Models
Zhang, Z., Zhang, A., Li, M., & Smola, A. (2022) · arXiv · doi:10.48550/arxiv.2210.03493 · Cited by 274 · see citing works
Towards Understanding Chain-of-Thought Prompting: An Empirical Study of What Matters
Wang, B., Min, S., Deng, X., Shen, J., Wu, Y., Zettlemoyer, L., & Sun, H. (2023) · Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2023.acl-long.153 · Cited by 120 · see citing works
Making Language Models Better Reasoners with Step-Aware Verifier
Li, Y., Lin, Z., Zhang, S., Fu, Q., Chen, B., Lou, J.-G., & Chen, W. (2023) · Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2023.acl-long.291 · Cited by 69 · see citing works
Lightman, H., Kosaraju, V., Burda, Y., Edwards, H., Baker, B., Lee, T., Leike, J., Schulman, J., Sutskever, I., & Cobbe, K. (2023) · arXiv · doi:10.48550/arxiv.2305.20050 · Cited by 65 · see citing works
QAScore—An Unsupervised Unreferenced Metric for the Question Generation Evaluation
Ji, T., Lyu, C., Jones, G., Zhou, L., & Graham, Y. (2022) · Entropy · doi:10.3390/e24111514 · Cited by 12 · see citing works
Flagged first, then by citations
Research works printed but not linked: 33
- Ekin Akyürek, Dale Schuurmans, Jacob Andreas, Tengyu Ma, and Denny Zhou. 2022. What learning algorithm is in-context learning? investigations with linear models. arXiv preprint arXiv:2211.15661. · arXiv
- Aida Amini, Saadia Gabriel, Peter Lin, Rik Koncel- Kedziorski, Yejin Choi, and Hannaneh Hajishirzi. 2019. Mathqa: Towards interpretable math word problem solving with operation-based formalisms. north american chapter of the association for com- putational linguistics.
- Tom Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared D Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, et al. 2020. Language models are few-shot learners. Advances in neural information processing systems, 33:1877-1901.
- Mark Chen, Jerry Tworek, Heewoo Jun, Qiming Yuan, Henrique Ponde de Oliveira Pinto, Jared Kaplan, Harrison Edwards, Yuri Burda, Nicholas Joseph, Greg Brockman, Alex Ray, Raul Puri, Gretchen Krueger, Michael Petrov, Heidy Khlaaf, Girish Sas- try, Pamela Mishkin, Brooke Chan, Scott Gray, Nick Ryder, Mikhail Pavlov, Alethea Power, Lukasz Kaiser, Mohammad Bavarian, Clemens Winter, Philippe Tillet, Felipe Petroski Such, Dave Cum- mings, Matthias Plappert, Fotios Chantzis, Eliza- beth Barnes, Ariel Herbert-Voss, William Hebgen Guss, Alex Nichol, Alex Paino, Nikolas Tezak, Jie Tang, Igor Babuschkin, Suchir Balaji, Shantanu Jain, William Saunders, Christopher Hesse, Andrew N. Carr, Jan Leike, Joshua Achiam, Vedant Misra, Evan Morikawa, Alec Radford, Matthew Knight, Miles Brundage, Mira Murati, Katie Mayer, Peter Welinder, Bob McGrew, Dario Amodei, Sam McCandlish, Ilya Sutskever, and Wojciech Zaremba. 2021. Evaluat- ing large language models trained on code. CoRR, abs/2107.03374. · arXiv
- Karl Cobbe, Vineet Kosaraju, Mohammad Bavar- ian, Jacob Hilton, Reiichiro Nakano, Christopher Hesse, and John Schulman. 2021. Training veri- fiers to solve math word problems. arXiv preprint arXiv:2110.14168. · arXiv
- Shehzaad Dhuliawala, Mojtaba Komeili, Jing Xu, Roberta Raileanu, Xian Li, Asli Celikyilmaz, and Jason Weston. 2023. Chain-of-verification reduces hallucination in large language models. ArXiv, abs/2309.11495. · arXiv
- Mor Geva, Daniel Khashabi, Elad Segal, Tushar Khot, Dan Roth, and Jonathan Berant. 2021. Did aristotle use a laptop? a question answering benchmark with implicit reasoning strategies. Transactions of the Association for Computational Linguistics.
- Yuxian Gu, Jiaxin Wen, Hao Sun, Yi Song, Pei Ke, Chujie Zheng, Zheng Zhang, Jianzhu Yao, Lei Liu, Xiaoyan Zhu, and Minlie Huang. 2023. Eva2.0: In- vestigating open-domain chinese dialogue systems with large-scale pre-training. Machine Intelligence Research, 20(2):207-219. · DOI
- Charles R Harris, K Jarrod Millman, Stéfan J Van Der Walt, Ralf Gommers, Pauli Virtanen, David Cour- napeau, Eric Wieser, Julian Taylor, Sebastian Berg, Nathaniel J Smith, et al. 2020. Array programming with numpy. Nature, 585(7825):357-362.
- Takeshi Kojima, Shixiang Shane Gu, Machel Reid, Yu- taka Matsuo, and Yusuke Iwasawa. 2022. Large lan- guage models are zero-shot reasoners. In Advances in Neural Information Processing Systems.
- Rik Koncel-Kedziorski, Hannaneh Hajishirzi, Ashish Sabharwal, Oren Etzioni, and Siena Dumas Ang. 2015. Parsing algebraic word problems into equa- tions. Transactions of the Association for Computa- tional Linguistics.
- Nate Kushman, Yoav Artzi, Luke Zettlemoyer, and Regina Barzilay. 2014. Learning to automatically solve algebra word problems. In Proceedings of the 52nd Annual Meeting of the Association for Compu- tational Linguistics (Volume 1: Long Papers), pages 271-281, Baltimore, Maryland. Association for Com- putational Linguistics. · DOI
- Bin Li, Encheng Chen, Hongru Liu, Yixuan Weng, Bin Sun, Shutao Li, Yongping Bai, and Meiling Hu. 2021. More but correct: Generating diversified and entity-revised medical response. arXiv preprint arXiv:2108.01266. · arXiv
- Yifei Li, Zeqi Lin, Shizhuo Zhang, Qiang Fu, Bei Chen, Jian-Guang Lou, and Weizhu Chen. 2022. On the advance of making language models better reasoners. arXiv preprint arXiv:2206.02336. · arXiv
- Arkil Patel, Satwik Bhattamishra, and Navin Goyal. 2021. Are nlp models really able to solve simple math word problems. north american chapter of the association for computational linguistics.
- David L Poole and Alan K Mackworth. 2010. Artificial Intelligence: foundations of computational agents. Cambridge University Press.
- Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, Ilya Sutskever, et al. 2019. Language models are unsupervised multitask learners. OpenAI blog, 1(8):9.
- Jack W. Rae, Sebastian Borgeaud, Trevor Cai, Katie Millican, Jordan Hoffmann, H. Francis Song, John Aslanides, Sarah Henderson, Roman Ring, Susan- nah Young, Eliza Rutherford, Tom Hennigan, Ja- cob Menick, Albin Cassirer, Richard Powell, George van den Driessche, Lisa Anne Hendricks, Mari- beth Rauh, Po-Sen Huang, Amelia Glaese, Jo- hannes Welbl, Sumanth Dathathri, Saffron Huang, Jonathan Uesato, John Mellor, Irina Higgins, Antonia Creswell, Nat McAleese, Amy Wu, Erich Elsen, Sid- dhant M. Jayakumar, Elena Buchatskaya, David Bud- den, Esme Sutherland, Karen Simonyan, Michela Pa- ganini, Laurent Sifre, Lena Martens, Xiang Lorraine Li, Adhiguna Kuncoro, Aida Nematzadeh, Elena Gribovskaya, Domenic Donato, Angeliki Lazaridou, Arthur Mensch, Jean-Baptiste Lespiau, Maria Tsim- poukelli, Nikolai Grigorev, Doug Fritz, Thibault Sot- tiaux, Mantas Pajarskas, Toby Pohlen, Zhitao Gong, Daniel Toyama, Cyprien de Masson d'Autume, Yujia Li, Tayfun Terzi, Vladimir Mikulik, Igor Babuschkin, Aidan Clark, Diego de Las Casas, Aurelia Guy, Chris Jones, James Bradbury, Matthew Johnson, Blake A. Hechtman, Laura Weidinger, Iason Gabriel, William S. Isaac, Edward Lockhart, Simon Osindero, Laura Rimell, Chris Dyer, Oriol Vinyals, Kareem Ayoub, Jeff Stanway, Lorrayne Bennett, Demis Hass- abis, Koray Kavukcuoglu, and Geoffrey Irving. 2021. Scaling language models: Methods, analysis & in- sights from training gopher. CoRR, abs/2112.11446. · arXiv
- Tian-Xiang Sun, Xiang-Yang Liu, Xi-Peng Qiu, and Xuan-Jing Huang. 2022. Paradigm shift in natural language processing. Machine Intelligence Research, 19(3):169-183. · DOI
- Xingwu Sun, Jing Liu, Yajuan Lyu, Wei He, Yanjun Ma, and Shi Wang. 2018. Answer-focused and position- aware neural question generation. In Proceedings of the 2018 Conference on Empirical Methods in Nat- ural Language Processing, pages 3930-3939, Brus- sels, Belgium. Association for Computational Lin- guistics. · DOI
- Alon Talmor, Jonathan Herzig, Nicholas Lourie, and Jonathan Berant. 2018. Commonsenseqa: A question answering challenge targeting commonsense knowl- edge. north american chapter of the association for computational linguistics.
- Siyuan Wang, Wanjun Zhong, Duyu Tang, Zhongyu Wei, Zhihao Fan, Daxin Jiang, Ming Zhou, and Nan Duan. 2022b. Logic-driven context extension and data augmentation for logical reasoning of text. In Findings of the Association for Computational Lin- guistics: ACL 2022, pages 1619-1629, Dublin, Ire- land. Association for Computational Linguistics. · DOI
- Xuezhi Wang, Jason Wei, Dale Schuurmans, Quoc V Le, Ed H. Chi, Sharan Narang, Aakanksha Chowdhery, and Denny Zhou. 2023b. Self-consistency improves chain of thought reasoning in language models. In The Eleventh International Conference on Learning Representations.
- Xuezhi Wang, Jason Wei, Dale Schuurmans, Quoc V Le, Ed H. Chi, Sharan Narang, Aakanksha Chowdhery, and Denny Zhou. 2023c. Self-consistency improves chain of thought reasoning in language models. In The Eleventh International Conference on Learning Representations.
- Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Ed Chi, Quoc Le, and Denny Zhou. 2022. Chain of thought prompting elicits reasoning in large language models. arXiv preprint arXiv:2201.11903. · arXiv
- Yixuan Weng, Bin Li, Fei Xia, Minjun Zhu, Bing Sun, Shizhu He, Kang Liu, and Jun Zhao. 2023a. Large language models need holistically thought in medical conversational qa. ArXiv, abs/2305.05410. · arXiv
- Fei Xia, Bin Li, Yixuan Weng, Shizhu He, Kang Liu, Bin Sun, Shutao Li, and Jun Zhao. 2022. Medconqa: Medical conversational question answering system based on knowledge graphs. In Conference on Em- pirical Methods in Natural Language Processing.
- Yisheng Xiao, Lijun Wu, Junliang Guo, Juntao Li, Min Zhang, Tao Qin, and Tie-yan Liu. 2022. A survey on non-autoregressive generation for neural machine translation and beyond. · DOI
- Ori Yoran, Alon Talmor, and Jonathan Berant. 2021. Turning tables: Generating examples from semi- structured tables for endowing language models with reasoning skills. arXiv: Computation and Language.
- Weihao Yu, Zihang Jiang, Yanfei Dong, and Jiashi Feng. 2020. Reclor: A reading comprehension dataset re- quiring logical reasoning. international conference on learning representations.
- Minjun Zhu, Bin Li, Yixuan Weng, and F. Xia. 2022a. A knowledge storage and semantic space alignment method for multi-documents dialogue generation. In Workshop on Document-grounded Dialogue and Con- versational Question Answering.
- Minjun Zhu, Yixuan Weng, Shizhu He, Kang Liu, and Jun Zhao. 2022b. Reasonchainqa: Text-based com- plex question answering with explainable evidence chains. ArXiv, abs/2210.08763. · arXiv
- Minjun Zhu, Yixuan Weng, Shizhu He, Cunguang Wang, Kang Liu, Li Cai, and Jun Zhao. 2023. Learn- ing to build reasoning chains by reliable path retrieval. In ICASSP 2023-2023 IEEE International Confer- ence on Acoustics, Speech and Signal Processing (ICASSP), pages 1-5. IEEE.
Cited by
Sorted by citations
Towards Reasoning in Large Language Models: A Survey
Huang, J., & Chang, K. C.-C. (2023) · Findings of the Association for Computational Linguistics: ACL 2023 · doi:10.18653/v1/2023.findings-acl.67 · Cited by 428 · see citing works
Plan-and-Solve Prompting: Improving Zero-Shot Chain-of-Thought Reasoning by Large Language Models
Wang, L., Xu, W., Lan, Y., Hu, Z., Lan, Y., Lee, R. K.-W., & Lim, E.-P. (2023) · Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2023.acl-long.147 · Cited by 344 · see citing works
Kamoi, R., Zhang, Y., Zhang, N., Han, J., & Zhang, R. (2024) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00713 · Cited by 82 · see citing works
Shojaee, P., Mirzadeh, I., Alizadeh vahid, K., Horton, M., Bengio, S., & Farajtabar, M. (2025) · Advances in Neural Information Processing Systems 38 · doi:10.52202/085713-3603 · Cited by 47 · see citing works
Zhang, Z., Yao, Y., Zhang, A., Tang, X., Ma, X., He, Z., Wang, Y., Gerstein, M., Wang, R., Liu, G., et al. (2025) · ACM Computing Surveys · doi:10.1145/3719341 · Cited by 51 · see citing works
TrueTeacher: Learning Factual Consistency Evaluation with Large Language Models
Gekhman, Z., Herzig, J., Aharoni, R., Elkind, C., & Szpektor, I. (2023) · Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2023.emnlp-main.127 · Cited by 33 · see citing works
LLMs cannot find reasoning errors, but can correct them given the error location
Tyen, G., Mansoor, H., Carbune, V., Chen, P., & Mak, T. (2024) · Findings of the Association for Computational Linguistics ACL 2024 · doi:10.18653/v1/2024.findings-acl.826 · Cited by 24 · see citing works
Why Do Multi-Agent LLM Systems Fail?
Cemri, M., Pan, M. Z., Yang, S., Agrawal, L. A., Chopra, B., Tiwari, R., Keutzer, K., Parameswaran, A., Klein, D., Ramchandran, K., et al. (2025) · arXiv · doi:10.48550/arxiv.2503.13657 · Cited by 15 · see citing works
“Interestingly, despite its stronger performance in FC1 and FC2, MetaGPT exhibits a considerably higher number of FC3 (Task Verification) failures than ChatDev.”
LLMs Can't Plan, But Can Help Planning in LLM-Modulo Frameworks
Kambhampati, S., Valmeekam, K., Guan, L., Verma, M., Stechly, K., Bhambri, S., Saldyt, L., & Murthy, A. (2024) · arXiv · doi:10.48550/arxiv.2402.01817 · Cited by 11 · see citing works
“…generate correct solutions in one go, their accuracy might improve in an iterative prompting regime, where LLMs will be able to ”self-critique” their candidate solutions and refine them to the point of correctness Yao et al. (2023b,a); Shinn et al. (2023); Weng et al. (2023); Huang et al. (2022).”
BackgroundEducationQ: Evaluating LLMs' Teaching Capabilities Through Multi-Agent Dialogue Framework
Shi, Y., Liang, R., & Xu, Y. (2025) · Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2025.acl-long.1576 · Cited by 11 · see citing works
“Markel et al. (2023) employed LLMs to simulate student dialogues for teacher training.”
A Survey of Post-Training Scaling in Large Language Models
Lai, H., Liu, X., Gao, J., Cheng, J., Qi, Z., Xu, Y., Yao, S., Zhang, D., Du, J., Hou, Z., et al. (2025) · Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2025.acl-long.140 · Cited by 8 · see citing works
Xie, E., Cheng, L., Shireman, J., Cai, Y., Liu, J., Mohanty, C., Dey, M., & Kendziorski, C. (2024) · doi:10.1101/2024.12.04.626476 · Cited by 3 · see citing works
On the Self-Verification Limitations of Large Language Models on Reasoning and Planning Tasks
Stechly, K., Valmeekam, K., & Kambhampati, S. (2024) · arXiv · doi:10.48550/arxiv.2402.08115 · Cited by 3 · see citing works
“These take many forms: common search algorithms with the LLM cast in the role of heuristic [37, 11], approaches which reduce error rates by enforcing various consistency conditions [8, 6, 17], and direct LLM self-critique [27, 35, 5, 15, 21, 38].”
BackgroundMethodDeepScientist: Advancing Frontier-Pushing Scientific Findings Progressively
Weng, Y., Zhu, M., Xie, Q., Sun, Q., Lin, Z., Liu, S., & Zhang, Y. (2025) · arXiv · doi:10.48550/arxiv.2509.26603
Self-citation
- Authors
- Weng, Y., Zhu, M., Xia, F., Li, B., He, S., Liu, S., Sun, B., Liu, K., & Zhao, J.
- Affiliations
- Chinese Academy of Sciences (CN)
- Hunan University (CN)
- Venue
- Findings of the Association for Computational Linguistics: EMNLP 2023
- Published
- 2023
- Pages
- 2550-2575
- OpenAlex
- W4389524159
- Cited by
- 110
- References
- 9
- Citations by year
- 2026: 31 · 2025: 42 · 2024: 26 · 2023: 11
- FWCI
- 19.49
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- Locations
- Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) (publishedVersion, cc-by)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- National Natural Science Foundation of China (61976211, U1936207, 62376270, 62171183)
- Chinese Academy of Sciences (XDA27020100)
- Youth Innovation Promotion Association of the Chinese Academy of Sciences
- Youth Innovation Promotion Association
- 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
- large language models, self-verification, chain-of-thought prompting, answer validation, reasoning, arithmetic reasoning, commonsense reasoning, logical reasoning, error accumulation
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
