When Can LLMs Actually Correct Their Own Mistakes? A Critical Survey of Self-Correction of LLMs
Open accessAbstract
Self-correction is an approach to improving responses from large language models (LLMs) by refining the responses using LLMs during inference. Prior work has proposed various self-correction frameworks using different sources of feedback, including self-evaluation and external feedback. However, there is still no consensus on the question of when LLMs can correct their own mistakes, as recent studies also report negative results. In this work, we critically survey broad papers and discuss the conditions required for successful self-correction. We first find that prior studies often do not define their research questions in detail and involve impractical frameworks or unfair evaluations that over-evaluate self-correction. To tackle these issues, we categorize research questions in self-correction research and provide a checklist for designing appropriate experiments. Our critical survey based on the newly categorized research questions shows that (1) no prior work demonstrates successful self-correction with feedback from prompted LLMs, except for studies in tasks that are exceptionally suited for self-correction, (2) self-correction works well in tasks that can use reliable external feedback, and (3) large-scale fine-tuning enables self-correction.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
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?
Surveys and targeted studies suggest that many gains attributed to reflection depend on external signals-tests, retrieval, or explicit error flags-with limited success at autonomous corrective reasoning when those signals are weak.
“Surveys and targeted studies, however, suggest that many gains attributed to reflection depend on external signals-tests, retrieval, or explicit error flags-with limited success at autonomous corrective reasoning when those signals are weak[3, 12-14].”
From Illusions of reflection: open-ended task reveals systematic failures in Large Language Models' reflective reasoning (2025) · cites this paper as [12]
No prior work demonstrates successful self-correction with feedback from prompted LLMs, except for studies in tasks exceptionally suited for self-correction.
“Kamoi et al. (2024), in a critical survey published in TACL, concluded that 'no prior work demonstrates successful self-correction with feedback from prompted LLMs, except for studies in tasks that are exceptionally suited for self-correction.'”
From Can LLMs Self-Correct Table Reasoning Errors? (2026) · cites this paper as Kamoi et al., 2024
LLMs struggle with self-correction without external guidance.
“This aligns with findings by Huang et al. (2023a); Kamoi et al. (2024), which showed that LLMs struggle with self-correction without external guidance.”
From LLM Self-Correction with DeCRIM: Decompose, Critique, and Refine for Enhanced Following of Instructions with Multiple Constraints (2024) · cites this paper as Kamoi et al. (2024)
The ability of an LLM to self-correct alone is limited to tasks where responses can be decomposed and rely on verifiable components.
“However, Kamoi et al. (2024); Huang et al. (2023a) demonstrated that the ability of LLM to self-correct alone is limited to tasks where responses can be decomposed and rely on verifiable components.”
From LLM Self-Correction with DeCRIM: Decompose, Critique, and Refine for Enhanced Following of Instructions with Multiple Constraints (2024) · cites this paper as Kamoi et al. (2024)
System 2 approaches like self-consistency and generate-then-rank do not directly refine responses and assume that LLMs can generate at least one correct initial response with a considerable probability, which is not always the case.
“These two approaches do not directly refine responses and assume that LLMs can generate at least one correct initial response with a considerable probability, which is not always the case.”
From LLM Self-Correction with DeCRIM: Decompose, Critique, and Refine for Enhanced Following of Instructions with Multiple Constraints (2024) · cites this paper as Kamoi et al. (2024)
For harder tasks, LLM self-correction may require additional modeling, new data, or even external tools.
“For harder tasks, LLM self-correction may require additional modeling, new data, or even external tools.”
From LLM Self-Correction with DeCRIM: Decompose, Critique, and Refine for Enhanced Following of Instructions with Multiple Constraints (2024) · cites this paper as Kamoi et al. (2024)
Self-correction has emerged as an effective approach for enhancing LLM responses during generation by refining them during generation time.
“Self-correction has emerged as an effective approach for enhancing LLM responses during generation by refining them during generation time (Pan et al., 2024; Kamoi et al., 2024).”
From LLM Self-Correction with DeCRIM: Decompose, Critique, and Refine for Enhanced Following of Instructions with Multiple Constraints (2024) · cites this paper as Kamoi et al., 2024
Iterative refinement replaces single-pass generation with multi-step processes where an initial output is progressively improved.
“This paradigm replaces the single-pass generation approach with multi-step processes where an initial output is progressively improved (Kamoi et al., 2024).”
From Can AI Be a Good Peer Reviewer? A Survey of Peer Review Process, Evaluation, and the Future (2026) · cites this paper as Kamoi et al., 2024
Iterative refinement is a powerful technique that mimics the human cognitive process of drafting, critiquing, and revising.
“Iterative refinement is a powerful technique that mimics the human cognitive process of drafting, critiquing, and revising (Xu et al., 2025;Kamoi et al., 2024).”
From Can AI Be a Good Peer Reviewer? A Survey of Peer Review Process, Evaluation, and the Future (2026) · cites this paper as Kamoi et al., 2024
Cites
Advances in neural information processing systems 7
(1996) · Computers & Mathematics with Applications · doi:10.1016/s0898-1221(96)90463-0 · Cited by 14,759 · see citing works
Proceedings of the 24th international conference on Machine learning
(2007) · doi:10.1145/1273496 · Cited by 11,869 · see citing works
Tree of Thoughts: Deliberate Problem Solving with Large Language Models
Yao, S., Yu, D., Zhao, J., Shafran, I., Griffiths, T. L., Cao, Y., & Narasimhan, K. (2023) · arXiv · doi:10.48550/arxiv.2305.10601 · Cited by 1,158 · see citing works
Reflexion: Language Agents with Verbal Reinforcement Learning
Shinn, N., Cassano, F., Berman, E., Gopinath, A., Narasimhan, K., & Yao, S. (2023) · arXiv · doi:10.48550/arxiv.2303.11366 · Cited by 1,087 · see citing works
Language (Technology) is Power: A Critical Survey of “Bias” in NLP
Blodgett, S. L., Barocas, S., Daumé III, H., & Wallach, H. (2020) · Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2020.acl-main.485 · Cited by 1,050 · cited 1× · see citing works
“…detection (Zhang et al., 2023c; Chern et al., 2023; Chen and Shu, 2024; Mishra et al., 2024), context-faithfulness (Wang et al., 2020; Durmus et al., 2020; Scialom et al., 2021), harmful content detection (Rauh et al., 2022), and bias detection (Blodgett et al., 2020; Feng et al., 2023).”
G-Eval: NLG Evaluation using Gpt-4 with Better Human Alignment
Liu, Y., Iter, D., Xu, Y., Wang, S., Xu, R., & Zhu, C. (2023) · Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2023.emnlp-main.153 · Cited by 948 · see citing works
G-Eval: NLG Evaluation using GPT-4 with Better Human Alignment
Liu, Y., Iter, D., Xu, Y., Wang, S., Xu, R., & Zhu, C. (2023) · arXiv · doi:10.48550/arxiv.2303.16634 · Cited by 948 · see citing works
Self-Consistency Improves Chain of Thought Reasoning in Language Models
Wang, X., Wei, J., Schuurmans, D., Le, Q., Chi, E., Narang, S., Chowdhery, A., & Zhou, D. (2022) · arXiv · doi:10.48550/arxiv.2203.11171 · Cited by 829 · see citing works
Self-Refine: Iterative Refinement with Self-Feedback
Madaan, A., Tandon, N., Gupta, P., Hallinan, S., Gao, L., Wiegreffe, S., Alon, U., Dziri, N., Prabhumoye, S., Yang, Y., et al. (2023) · Advances in Neural Information Processing Systems 36 · doi:10.52202/075280-2019 · Cited by 755 · see citing works
SequenceR: Sequence-to-Sequence Learning for End-to-End Program Repair
Chen, Z.-M., Kommrusch, S., Tufano, M., Pouchet, L., Poshyvanyk, D., & Martin, M. (2018) · IEEE Transactions on Software Engineering · doi:10.1109/tse.2019.2940179 · Cited by 585 · see citing works
The CoNLL-2014 Shared Task on Grammatical Error Correction
Ng, H. T., Wu, S. M., Briscoe, T., Hadiwinoto, C., Susanto, R. H., & Bryant, C. (2014) · doi:10.3115/v1/w14-1701 · Cited by 578 · cited 1× · see citing works
“– Self-correction works in many tasks when large training data for feedback generation is available – Tasks that can use reinforcement learning or self-corrective learning (Welleck et al., 2023), i.e., tasks whose responses can be easily evaluated given ground-truth answers 8 Checklist for…”
Training Verifiers to Solve Math Word Problems
Cobbe, K., Kosaraju, V., Bavarian, M., Chen, M., Jun, H., Kaiser, L., Plappert, M., Tworek, J., Hilton, J., Nakano, R., et al. (2021) · arXiv · doi:10.48550/arxiv.2110.14168 · Cited by 575 · see citing works
SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Models
Manakul, P., Liusie, A., & Gales, M. (2023) · Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2023.emnlp-main.557 · Cited by 557 · see citing works
Active Retrieval Augmented Generation
Jiang, Z., Xu, F., Gao, L., Sun, Z., Liu, Q., Dwivedi-Yu, J., Yang, Y., Callan, J., & Neubig, G. (2023) · Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2023.emnlp-main.495 · Cited by 497 · see citing works
Socratic guiding questions in synthetic arithmetic data: matched LoRA runs (revision v2)
O'Grady, J., Gürhan, A. I., Chee, F. T., et al. (2026) · Zenodo (CERN European Organization for Nuclear Research) · doi:10.5281/zenodo.23236017 · Cited by 463 · see citing works
DeepFix: Fixing Common C Language Errors by Deep Learning
Gupta, R., Pal, S., Kanade, A., & Shevade, S. (2017) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v31i1.10742 · Cited by 426 · cited 1× · see citing works
“…Schick et al., 2023), grammatical error correction (Ng et al., 2014; Lichtarge et al., 2019), factual error correction (Cao et al., 2020; Thorne and Vlachos, 2021), and code repair (Gupta et al., 2017; Mesbah et al., 2019; Bader et al., 2019; Chen et al., 2021; Yasunaga and Liang, 2020, 2021).”
FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generation
Min, S., Krishna, K., Lyu, X., Lewis, M., Yih, W.-T., Koh, P., Iyyer, M., Zettlemoyer, L., & Hajishirzi, H. (2023) · Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2023.emnlp-main.741 · Cited by 407 · see citing works
Analyzing and Mitigating Object Hallucination in Large Vision-Language ModelsPreprint
Zhou, Y.-Y., Cui, C.-H., Yoon, J., Zhang, L.-J., Deng, Z., Finn, C., Bansal, M., & Yao, H. (2023) · International Conference on Learning Representations · doi:10.48550/arxiv.2310.00754 · Cited by 399 · see citing works
(2020) · doi:10.18653/v1/2020.acl-main.450 · Cited by 356 · cited 1× · see citing works
“…including misinformation detection (Zhang et al., 2023c; Chern et al., 2023; Chen and Shu, 2024; Mishra et al., 2024), context-faithfulness (Wang et al., 2020; Durmus et al., 2020; Scialom et al., 2021), harmful content detection (Rauh et al., 2022), and bias detection (Blodgett et al.,…”
LLM Self Defense: By Self Examination, LLMs Know They Are Being TrickedPreprint
Helbling, A., Phute, M., Hull, M., Szyller, S., & Chau, D. H. (2023) · Tiny Papers @ ICLR · doi:10.48550/arxiv.2308.07308 · Cited by 351 · cited 1× · see citing works
“…2023; Ki and Carpuat, 2024), information retrieval (Gero et al., 2023), vision language tasks (Yin et al., 2023; Ge et al., 2023; Zhou et al., 2024; Lee et al., 2024; Huang et al., 2024b; Liu et al., 2024), and prompt optimization (Pryzant et al., 2023; Mehrabi et al., 2023; Yang et al., 2024).”
Can Large Language Models Be an Alternative to Human Evaluations?
Chiang, C.-H., & Lee, H.-Y. (2023) · Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2023.acl-long.870 · Cited by 331 · see citing works
Bai, Y., Kadavath, S., Kundu, S., Askell, A., Kernion, J., Jones, A., Chen, A., Goldie, A., Mirhoseini, A., & Kaplan, J. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2212.08073 · Cited by 326 · see citing works
Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debate
Liang, T., He, Z., Jiao, W., Wang, X., Wang, Y., Wang, R., Yang, Y., Shi, S., & Tu, Z. (2023) · arXiv · doi:10.48550/arxiv.2305.19118 · Cited by 320 · see citing works
Durmus, E., He, H., & Diab, M. (2020) · Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2020.acl-main.454 · Cited by 306 · see citing works
Exploring Collaboration Mechanisms for LLM Agents: A Social Psychology ViewPreprint
Zhang, J., Xu, X., Liu, R., & Deng, S.-M. (2023) · arXiv.org · doi:10.48550/arxiv.2310.02124 · Cited by 300 · see citing works
Chain-of-Verification Reduces Hallucination in Large Language Models
Dhuliawala, S., Komeili, M., Xu, J., Raileanu, R., Li, X., Celikyilmaz, A., & Weston, J. (2024) · Findings of the Association for Computational Linguistics ACL 2024 · doi:10.18653/v1/2024.findings-acl.212 · Cited by 282 · see citing works
Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debate
Liang, T., He, Z., Jiao, W., Wang, X., Wang, Y., Wang, R., Yang, Y., Shi, S., & Tu, Z. (2024) · Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2024.emnlp-main.992 · Cited by 274 · see citing works
Getafix: learning to fix bugs automatically
Bader, J., Scott, A., Pradel, M., & Chandra, S. (2019) · Proceedings of the ACM on Programming Languages · doi:10.1145/3360585 · Cited by 238 · cited 1× · see citing works
“…Schick et al., 2023), grammatical error correction (Ng et al., 2014; Lichtarge et al., 2019), factual error correction (Cao et al., 2020; Thorne and Vlachos, 2021), and code repair (Gupta et al., 2017; Mesbah et al., 2019; Bader et al., 2019; Chen et al., 2021; Yasunaga and Liang, 2020, 2021).”
Large Language Models Can Self-Improve
Huang, J., Gu, S., Hou, L., Wu, Y., Wang, X., Yu, H., & Han, J. (2023) · Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2023.emnlp-main.67 · Cited by 230 · see citing works
Self-citationLarge Language Models Can Self-Improve
Huang, J., Gu, S. S., Hou, L., Wu, Y., Wang, X., Yu, H., & Han, J. (2022) · arXiv · doi:10.48550/arxiv.2210.11610 · Cited by 230 · see citing works
Self-citationSelf-Diagnosis and Self-Debiasing: A Proposal for Reducing Corpus-Based Bias in NLP
Schick, T., Udupa, S., & Schütze, H. (2021) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00434 · Cited by 227 · cited 1× · see citing works
“…translation (Chen et al., 2023b; Raunak et al., 2023; Ki and Carpuat, 2024), information retrieval (Gero et al., 2023), vision language tasks (Yin et al., 2023; Ge et al., 2023; Zhou et al., 2024; Lee et al., 2024; Huang et al., 2024b; Liu et al., 2024), and prompt optimization (Pryzant et al.,…”
Woodpecker: hallucination correction for multimodal large language models
Yin, S., Fu, C., Zhao, S., Xu, T. B., Wang, H., Sui, D., Shen, Y., Li, K., Sun, X., & Chen, E. (2024) · Science China Information Sciences · doi:10.1007/s11432-024-4251-x · Cited by 196 · see citing works
Automatic Prompt Optimization with “Gradient Descent” and Beam Search
Pryzant, R., Iter, D., Li, J., Lee, Y., Zhu, C., & Zeng, M. (2023) · Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2023.emnlp-main.494 · Cited by 191 · see citing works
(2022) · doi:10.52202/068431-1126 · Cited by 182 · see citing works
QuestEval: Summarization Asks for Fact-based Evaluation
Scialom, T., Dray, P.-A., Lamprier, S., Piwowarski, B., Staiano, J., Wang, A., & Gallinari, P. (2021) · Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2021.emnlp-main.529 · Cited by 175 · see citing works
Reasoning with Language Model is Planning with World Model
Hao, S., Gu, Y., Ma, H., Hong, J., Wang, Z., Wang, D., & Hu, Z. (2023) · Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2023.emnlp-main.507 · Cited by 168 · see citing works
Peng, B., Galley, M., He, P., Cheng, H., Xie, Y., Hu, Y., Huang, Q., Liden, L., Yu, Z., Chen, W., et al. (2023) · arXiv · doi:10.48550/arxiv.2302.12813 · Cited by 156 · see citing works
Large Language Models as Optimizers
Yang, C., Wang, X., Lu, Y., Liu, H., Le, Q. V., Zhou, D., & Chen, X. (2023) · arXiv · doi:10.48550/arxiv.2309.03409 · Cited by 151 · see citing works
Feng, S., Park, C. Y., Liu, Y., & Tsvetkov, Y. (2023) · Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2023.acl-long.656 · Cited by 149 · see citing works
Logic-LM: Empowering Large Language Models with Symbolic Solvers for Faithful Logical Reasoning
Pan, L., Albalak, A., Wang, X., & Wang, W. Y. (2023) · arXiv · doi:10.48550/arxiv.2305.12295 · Cited by 144 · see citing works
Coder Reviewer Reranking for Code GenerationPreprint
Zhang, T., Yu, T., Hashimoto, T. B., Lewis, M., Yih, W.-T., Fried, D., & Wang, S. I. (2022) · International Conference on Machine Learning · doi:10.48550/arxiv.2211.16490 · Cited by 144 · see citing works
RARR: Researching and revising what language models say, using language models
Gao, L., Dai, Z., Pasupat, P., Chen, A., Chaganty, A. T., Fan, Y., Zhao, V., Lao, N., Lee, H., Juan, D.-C., et al. (2023) · Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2023.acl-long.910 · Cited by 141 · see citing works
Corpora Generation for Grammatical Error Correction
Lichtarge, J., Alberti, C., Kumar, S., et al. (2019) · Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT) · doi:10.18653/v1/n19-1333 · Cited by 140 · cited 1× · see citing works
“…domains, including information update (Shah et al., 2020; Iv et al., 2022; Schick et al., 2023), grammatical error correction (Ng et al., 2014; Lichtarge et al., 2019), factual error correction (Cao et al., 2020; Thorne and Vlachos, 2021), and code repair (Gupta et al., 2017; Mesbah et al.,…”
Language Models can Solve Computer Tasks
Kim, G., Baldi, P., & McAleer, S. (2023) · Advances in Neural Information Processing Systems 36 · doi:10.52202/075280-1723 · Cited by 137 · see citing works
Can LLM-Generated Misinformation Be Detected?Preprint
Chen, C., & Shu, K. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2309.13788 · Cited by 135 · see citing works
Factual Error Correction for Abstractive Summarization Models
Cao, M., Dong, Y., Wu, J., & Cheung, J. C. K. (2020) · Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2020.emnlp-main.506 · Cited by 132 · see citing works
Varshney, N., Yao, W., Zhang, H., Chen, J., & Yu, D. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2307.03987 · Cited by 128 · see citing works
Maieutic Prompting: Logically Consistent Reasoning with Recursive Explanations
Jung, J., Qin, L., Welleck, S., Brahman, F., Bhagavatula, C., Bras, R. J. L., & Choi, Y. (2022) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2022.emnlp-main.82 · Cited by 117 · see citing works
Large Language Models are Better Reasoners with Self-Verification
Weng, Y., Zhu, M., Xia, F., Li, B., He, S., Liu, S., Sun, B., Liu, K., & Zhao, J. (2023) · Findings of the Association for Computational Linguistics: EMNLP 2023 · doi:10.18653/v1/2023.findings-emnlp.167 · Cited by 110 · see citing works
DeepDelta: learning to repair compilation errors
Mesbah, A., Rice, A. C., Johnston, E., Glorioso, N., & Aftandilian, E. E. (2019) · ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering (ESEC/FSE) · doi:10.1145/3338906.3340455 · Cited by 109 · cited 1× · see citing works
“…Schick et al., 2023), grammatical error correction (Ng et al., 2014; Lichtarge et al., 2019), factual error correction (Cao et al., 2020; Thorne and Vlachos, 2021), and code repair (Gupta et al., 2017; Mesbah et al., 2019; Bader et al., 2019; Chen et al., 2021; Yasunaga and Liang, 2020, 2021).”
FLIRT: Feedback Loop In-context Red TeamingPreprint
Mehrabi, N., Goyal, P., Dupuy, C., et al. (2023) · Conference on Empirical Methods in Natural Language Processing · doi:10.48550/arxiv.2308.04265 · Cited by 103 · see citing works
Improving Factuality and Reasoning in Language Models through Multiagent Debate
Du, Y., Li, S., Torralba, A., Tenenbaum, J. B., & Mordatch, I. (2023) · arXiv · doi:10.48550/arxiv.2305.14325 · Cited by 96 · see citing works
Language Model Self-improvement by Reinforcement Learning ContemplationPreprint
Pang, J.-C., Wang, P.-Y., Li, K., Chen, X.-H., Xu, J., Zhang, Z.-Z., & Yu, Y. (2023) · International Conference on Learning Representations · doi:10.48550/arxiv.2305.14483 · Cited by 95 · see citing works
Baldur: Whole-Proof Generation and Repair with Large Language Models
First, E., Rabe, M. N., Ringer, T., & Brun, Y. (2023) · Proceedings of the 31st ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering · doi:10.1145/3611643.3616243 · Cited by 94 · see citing works
Re3: Generating Longer Stories With Recursive Reprompting and Revision
Yang, K., Tian, Y., Peng, N., & Klein, D. (2022) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2022.emnlp-main.296 · Cited by 91 · see citing works
Re3: Generating Longer Stories With Recursive Reprompting and Revision
Yang, K., Tian, Y., Peng, N., & Klein, D. (2022) · arXiv · doi:10.48550/arxiv.2210.06774 · Cited by 91 · see citing works
Generate & Rank: A Multi-task Framework for Math Word Problems
Shen, J., Yin, Y., Li, L., Shang, L., Jiang, X., Zhang, M., & Liu, Q. (2021) · Findings of the Association for Computational Linguistics: EMNLP 2021 · doi:10.18653/v1/2021.findings-emnlp.195 · Cited by 88 · cited 2× · see citing works
“…approach ranks responses using self-evaluation (Weng et al., 2023; Zhang et al., 2023d), confidence (Manakul et al., 2023), fine-tuned verifiers (Cobbe et al., 2021; Shen et al., 2021; Lightman et al., 2024), or verifiers with external tools (Shi et al., 2022; Chen et al., 2023a; Ni et al., 2023).”
REFINER: Reasoning Feedback on Intermediate Representations
Paul, D., Ismayilzada, M., Peyrard, M., Borges, B., Bosselut, A., West, R., & Faltings, B. V. (2024) · Conference of the European Chapter of the Association for Computational Linguistics · doi:10.18653/v1/2024.eacl-long.67 · Cited by 88 · see citing works
Pan, L., Saxon, M., Xu, W., Nathani, D., Wang, X., & Wang, W. Y. (2024) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00660 · Cited by 86 · see citing works
Graph-based, Self-Supervised Program Repair from Diagnostic Feedback
Yasunaga, M., & Liang, P. (2020) · arXiv · doi:10.48550/arxiv.2005.10636 · Cited by 81 · see citing works
Rethinking the Bounds of LLM Reasoning: Are Multi-Agent Discussions the Key?
Wang, Q., Wang, Z., Su, Y., Tong, H., & Song, Y. (2024) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2024.acl-long.331 · Cited by 80 · see citing works
How Language Model Hallucinations Can Snowball
Zhang, M., Press, O., Merrill, W., Liu, A., & Smith, N. A. (2023) · arXiv · doi:10.48550/arxiv.2305.13534 · Cited by 77 · see citing works
Teaching Large Language Models to Self-Debug
Chen, X., Lin, M., Schärli, N., & Zhou, D. (2023) · arXiv · doi:10.48550/arxiv.2304.05128 · Cited by 77 · see citing works
Natural Language to Code Translation with Execution
Shi, F., Fried, D., Ghazvininejad, M., Zettlemoyer, L., & Wang, S. I. (2022) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2022.emnlp-main.231 · Cited by 65 · 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
Exploring Collaboration Mechanisms for LLM Agents: A Social Psychology View
Zhang, J., Xu, X., Zhang, N., Liu, R., Hooi, B., & Deng, S. (2024) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2024.acl-long.782 · Cited by 64 · see citing works
Self-citationReConcile: Round-Table Conference Improves Reasoning via Consensus among Diverse LLMs
Chen, J., Saha, S., & Bansal, M. (2024) · Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2024.acl-long.381 · Cited by 64 · see citing works
ReConcile: Round-Table Conference Improves Reasoning via Consensus among Diverse LLMs
Chen, J. C.-Y., Saha, S., & Bansal, M. (2023) · arXiv · doi:10.48550/arxiv.2309.13007 · Cited by 64 · see citing works
Self-Edit: Fault-Aware Code Editor for Code Generation
Zhang, K., Li, Z., Li, J., Li, G., & Jin, Z. G. (2023) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2023.acl-long.45 · Cited by 61 · see citing works
LM vs LM: Detecting Factual Errors via Cross Examination
Cohen, R., Hamri, M., Geva, M., & Globerson, A. (2023) · Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2023.emnlp-main.778 · Cited by 60 · see citing works
Verify-and-Edit: A Knowledge-Enhanced Chain-of-Thought Framework
Zhao, R., Li, X., Joty, S., Qin, C., & Bing, L. (2023) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2023.acl-long.320 · Cited by 59 · see citing works
INSTRUCTSCORE: Towards Explainable Text Generation Evaluation with Automatic Feedback
Xu, W., Wang, D., Pan, L., Song, Z., Freitag, M., Wang, W. Y., & Li, L. (2023) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2023.emnlp-main.365 · Cited by 57 · see citing works
CRITIC: Large Language Models Can Self-Correct with Tool-Interactive Critiquing
Gou, Z., Shao, Z., Gong, Y., Shen, Y., Yang, Y., Duan, N., & Chen, W. (2023) · arXiv · doi:10.48550/arxiv.2305.11738 · Cited by 57 · see citing works
ChatEval: Towards Better LLM-based Evaluators through Multi-Agent Debate
Chan, C.-M., Chen, W., Su, Y., Yu, J., Xue, W., Zhang, S., Fu, J., & Liu, Z. (2023) · arXiv · doi:10.48550/arxiv.2308.07201 · Cited by 55 · see citing works
Self-critiquing models for assisting human evaluators
Saunders, W., Yeh, C., Wu, J., Bills, S., Ouyang, L., Ward, J., & Leike, J. (2022) · arXiv · doi:10.48550/arxiv.2206.05802 · Cited by 50 · see citing works
Fine-grained Hallucination Detection and Editing for Language Models
Mishra, A., Asai, A., Balachandran, V., Wang, Y., Neubig, G., Tsvetkov, Y., & Hajishirzi, H. (2024) · arXiv · doi:10.48550/arxiv.2401.06855 · Cited by 49 · see citing works
PEER: A Collaborative Language Model
Schick, T., Dwivedi-Yu, J., Jiang, Z., Petroni, F., Lewis, P., Izacard, G., You, Q., Nalmpantis, C., Grave, E., & Riedel, S. (2022) · arXiv · doi:10.48550/arxiv.2208.11663 · Cited by 49 · see citing works
LEVER: Learning to Verify Language-to-Code Generation with Execution
Ni, A., Iyer, S., Radev, D., Stoyanov, V., Yih, W.-T., Wang, S. I., & Lin, X. V. (2023) · arXiv · doi:10.48550/arxiv.2302.08468 · Cited by 46 · see citing works
RL4F: Generating Natural Language Feedback with Reinforcement Learning for Repairing Model Outputs
Akyürek, A. F., Akyürek, E., Kalyan, A., Clark, P. E., Wijaya, D. T., & Tandon, N. (2023) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2023.acl-long.427 · Cited by 45 · see citing works
Leveraging GPT-4 for Automatic Translation Post-Editing
Raunak, V., Sharaf, A., Wang, Y., Awadalla, H. H., & Menezes, A. (2023) · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) · doi:10.18653/v1/2023.findings-emnlp.804 · Cited by 45 · see citing works
Break-It-Fix-It: Unsupervised Learning for Program RepairPreprint
Yasunaga, M., & Liang, P. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2106.06600 · Cited by 43 · see citing works
Chain-of-Verification Reduces Hallucination in Large Language Models
Dhuliawala, S., Komeili, M., Xu, J., Raileanu, R., Li, X., Celikyilmaz, A., & Weston, J. (2023) · arXiv · doi:10.48550/arxiv.2309.11495 · Cited by 42 · see citing works
Generating Sequences by Learning to Self-Correct
Welleck, S., Lu, X., West, P., Brahman, F., Shen, T., Khashabi, D., & Choi, Y. (2022) · arXiv · doi:10.48550/arxiv.2211.00053 · Cited by 41 · see citing works
Iterative Translation Refinement with Large Language ModelsPreprint
Chen, P., Guo, Z., Haddow, B., et al. (2023) · European Association for Machine Translation Conferences/Workshops · doi:10.48550/arxiv.2306.03856 · Cited by 40 · see citing works
Evidence-based Factual Error Correction
Thorne, J., & Vlachos, A. (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.256 · Cited by 39 · see citing works
Faithful Reasoning Using Large Language Models
Creswell, A., & Shanahan, M. (2022) · arXiv · doi:10.48550/arxiv.2208.14271 · Cited by 37 · cited 1× · see citing works
“…response for each reasoning step using generation probability (Hao et al., 2023; Tyen et al., 2024), prompted self-evaluation (Jung et al., 2022; Creswell and Shanahan, 2022; Xie et al., 2023; Yao et al., 2023; Miao et al., 2024), or fine-tuned verifiers (Uesato et al., 2022; Tafjord et al.,…”
Large Language Models Cannot Self-Correct Reasoning Yet
Huang, J., Chen, X., Mishra, S., Zheng, H. S., Yu, A. W., Song, X., & Zhou, D. (2023) · arXiv · doi:10.48550/arxiv.2310.01798 · Cited by 36 · see citing works
Branch-Solve-Merge Improves Large Language Model Evaluation and Generation
Saha, S., Levy, O., Çelikyılmaz, A., Bansal, M., Weston, J., & Li, X. (2024) · Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT) · doi:10.18653/v1/2024.naacl-long.462 · Cited by 32 · see citing works
Entailer: Answering Questions with Faithful and Truthful Chains of Reasoning
Tafjord, O., Mishra, B. D., & Clark, P. A. (2022) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2022.emnlp-main.134 · Cited by 31 · see citing works
Tihanyi, N., Jain, R., Charalambous, Y., et al. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2305.14752 · Cited by 30 · cited 1× · see citing works
“…including Reasoning: arithmetic reasoning (Madaan et al., 2023; Nathani et al., 2023; Gou et al., 2024), code generation (Jiang et al., 2023a; Charalambous et al., 2023; Gou et al., 2024; Chen et al., 2024d; Olausson et al., 2024), proof generation (First et al., 2023), logical reasoning…”
Chern, I.-C., Chern, S., Chen, S., Yuan, W., Feng, K., Zhou, C., He, J., Neubig, G., & Liu, P. (2023) · arXiv · doi:10.48550/arxiv.2307.13528 · Cited by 29 · see citing works
Generating Natural Language Proofs with Verifier-Guided Search
Yang, K., Deng, J., & Chen, D. (2022) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2022.emnlp-main.7 · Cited by 29 · see citing works
Self-Evaluation Guided Beam Search for Reasoning
Xie, Y., Kawaguchi, K., Zhao, Y., Zhao, J. X., Kan, M.-Y., He, J., & Xie, M. (2023) · Advances in Neural Information Processing Systems 36 · doi:10.52202/075280-1802 · Cited by 27
Solving math word problems with process- and outcome-based feedback
Uesato, J., Kushman, N., Kumar, R., Song, F., Siegel, N., Wang, L., Creswell, A., Irving, G., & Higgins, I. (2022) · arXiv · doi:10.48550/arxiv.2211.14275 · Cited by 25 · 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
On the Intrinsic Self-Correction Capability of LLMs: Uncertainty and Latent ConceptPreprint
Liu, G.-D., Mao, H.-T., Cao, B., Xue, Z.-Y., Johnson, K., Tang, J.-L., & Wang, R.-R. (2024) · arXiv.org · doi:10.48550/arxiv.2406.02378 · Cited by 21 · see citing works
Self-Verification Improves Few-Shot Clinical Information Extraction
Gero, Z., Singh, C., Cheng, H., Naumann, T., Galley, M., Gao, J., & Poon, H. (2023) · arXiv · doi:10.48550/arxiv.2306.00024 · Cited by 21 · see citing works
Reinforced Self-Training (ReST) for Language Modeling
Gulcehre, C., Paine, T. L., Srinivasan, S., Konyushkova, K., Weerts, L., Sharma, A., Siddhant, A., Ahern, A., Wang, M., Gu, C., et al. (2023) · arXiv · doi:10.48550/arxiv.2308.08998 · Cited by 20 · see citing works
Volcano: Mitigating Multimodal Hallucination through Self-Feedback Guided Revision
Lee, S., Park, S. H., Jo, Y., & Seo, M. (2024) · Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers) · doi:10.18653/v1/2024.naacl-long.23 · Cited by 20 · see citing works
SelfEvolve: A Code Evolution Framework via Large Language ModelsPreprint
Jiang, S., Wang, Y., & Wang, Y. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2306.02907 · Cited by 18 · see citing works
SelfCheck: Using LLMs to Zero-Shot Check Their Own Step-by-Step Reasoning
Miao, N., Teh, Y. W., & Rainforth, T. (2023) · arXiv · doi:10.48550/arxiv.2308.00436 · Cited by 18 · see citing works
Characteristics of Harmful Text: Towards Rigorous Benchmarking of Language Models
Rauh, M., Mellor, J., Uesato, J., Huang, P.-S., Welbl, J., Weidinger, L., Dathathri, S., Glaese, A., Irving, G., Gabriel, I., et al. (2022) · neural information processing systems · doi:10.52202/068431-1793 · Cited by 15 · see citing works
FRUIT: Faithfully Reflecting Updated Information in Text
Iv, R., Passos, A., Singh, S., & Chang, M.-W. (2022) · Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT) · doi:10.18653/v1/2022.naacl-main.269 · Cited by 14 · cited 1× · see citing works
“To tackle these issues, we provide a checklist for self-correction research that provides requirements for designing appropriate experiments for verifying target RQs and recommended experiments for comprehensive analysis.”
Improving Language Models via Plug-and-Play Retrieval Feedback
Yu, W., Zhang, Z., Liang, Z., Jiang, M., & Sabharwal, A. (2023) · arXiv · doi:10.48550/arxiv.2305.14002 · Cited by 14 · see citing works
Small Language Model Can Self-Correct
Han, H., Liang, J., Shi, J., He, Q., & Xiao, Y. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i16.29774 · Cited by 13 · see citing works
Guiding Large Language Models to Post-Edit Machine Translation with Error Annotations
Ki, D., & Carpuat, M. J. (2024) · Findings of the Association for Computational Linguistics: NAACL · doi:10.18653/v1/2024.findings-naacl.265 · Cited by 13 · see citing works
Is Self-Repair a Silver Bullet for Code Generation?
Olausson, T. X., Inala, J. P., Wang, C., Gao, J., & Solar-Lezama, A. (2023) · arXiv · doi:10.48550/arxiv.2306.09896 · Cited by 13 · see citing works
Large Language Models Can Self-Correct with Minimal EffortPreprint
Wu, Z.-Y., Zeng, Q.-K., Zhang, Z., Tan, Z.-X., Shen, C., & Jiang, M. (2024) · arXiv.org · doi:10.48550/arxiv.2405.14092 · Cited by 12 · see citing works
When Hindsight is Not 20/20: Testing Limits on Reflective Thinking in Large Language Models
Li, Y., Yang, C., & Ettinger, A. (2024) · Findings of the Association for Computational Linguistics: NAACL · doi:10.18653/v1/2024.findings-naacl.237 · Cited by 11 · see citing works
A Closer Look at the Self-Verification Abilities of Large Language Models in Logical Reasoning
Hong, R., Zhang, H., Pang, X., Yu, D., & Zhang, C. (2024) · Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers) · doi:10.18653/v1/2024.naacl-long.52 · Cited by 11 · see citing works
Do LVLMs Understand Charts? Analyzing and Correcting Factual Errors in Chart Captioning
Huang, K.-H., Zhou, M., Chan, H. P., et al. (2024) · doi:10.18653/v1/2024.findings-acl.41 · Cited by 11 · see citing works
A Theoretical Understanding of Self-Correction through In-context Alignment
Wang, Y., Wu, Y., Wei, Z., Jegelka, S., & Wang, Y. (2024) · Advances in Neural Information Processing Systems 37 · doi:10.52202/079017-2854 · Cited by 11 · see citing works
Training Language Models with Language Feedback
Scheurer, J., Campos, J. A., Chan, J. S., Chen, A., Cho, K., & Perez, E. (2022) · arXiv · doi:10.48550/arxiv.2204.14146 · Cited by 11 · see citing works
GPT-4 Doesn't Know It's Wrong: An Analysis of Iterative Prompting for Reasoning Problems
Stechly, K., Marquez, M., & Kambhampati, S. (2023) · arXiv · doi:10.48550/arxiv.2310.12397 · Cited by 11 · see citing works
Generating Training Data with Language Models: Towards Zero-Shot Language Understanding
Meng, Y., Huang, J., Zhang, Y. B., & Han, J. (2022) · neural information processing systems · doi:10.52202/068431-0034 · Cited by 9 · see citing works
Self-citationFLIRT: Feedback Loop In-context Red Teaming
Mehrabi, N., Goyal, P., Dupuy, C., et al. (2024) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2024.emnlp-main.41 · Cited by 8 · see citing works
Halo: Estimation and Reduction of Hallucinations in Open-Source Weak Large Language ModelsPreprint
Elaraby, M., Lu, M., Dunn, J., et al. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2308.11764 · Cited by 7 · see citing works
SELF-[IN]CORRECT: LLMs Struggle with Discriminating Self-Generated Responses
Jiang, D., Zhang, J., Weller, O., Weir, N., Van Durme, B., & Khashabi, D. (2024) · arXiv · doi:10.48550/arxiv.2404.04298 · Cited by 7 · see citing works
SELF-[IN]CORRECT: LLMs Struggle with Discriminating Self-Generated Responses
Jiang, D., Zhang, J., Weller, O., Weir, N., Van Durme, B., & Khashabi, D. (2025) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v39i23.34603 · Cited by 7
MAF: Multi-Aspect Feedback for Improving Reasoning in Large Language Models
Nathani, D., Wang, D., Pan, L., & Wang, W. Y. (2023) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2023.emnlp-main.407 · Cited by 6 · see citing works
Small Language Models Need Strong Verifiers to Self-Correct Reasoning
Zhang, Y., Khalifa, M., Logeswaran, L., Kim, J., Lee, M., Lee, H., & Wang, L. (2024) · Findings of the Association for Computational Linguistics ACL 2024 · doi:10.18653/v1/2024.findings-acl.924 · Cited by 6 · see citing works
Self-citationEvaluating LLMs at Detecting Errors in LLM Responses
Kamoi, R., Das, S. S. S., Lou, R., Ahn, J. J., Zhao, Y., Lu, X., Zhang, N., Zhang, Y., Zhang, R. H., Vummanthala, S. R., et al. (2024) · arXiv · doi:10.48550/arxiv.2404.03602 · Cited by 5 · see citing works
Self-citationFrom Wrong To Right: A Recursive Approach Towards Vision-Language Explanation
Ge, J., Subramanian, S., Darrell, T., & Li, B. (2023) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2023.emnlp-main.75 · Cited by 4 · see citing works
Confidence Matters: Revisiting Intrinsic Self-Correction Capabilities of Large Language ModelsPreprint
Li, L., Chen, Z., Chen, G., et al. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2402.12563 · Cited by 4 · cited 1× · see citing works
“…al., 2022; Schick et al., 2023), grammatical error correction (Ng et al., 2014; Lichtarge et al., 2019), factual error correction (Cao et al., 2020; Thorne and Vlachos, 2021), and code repair (Gupta et al., 2017; Mesbah et al., 2019; Bader et al., 2019; Chen et al., 2021; Yasunaga and Liang,…”
Self-citationAutomatic Fact-Guided Sentence Modification
Shah, D., Schuster, T., & Barzilay, R. (2020) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v34i05.6406 · Cited by 2 · cited 1× · see citing works
“Editing Human-Written Text. by using language models has been studied in various domains, including information update (Shah et al., 2020; Iv et al., 2022; Schick et al., 2023), grammatical error correction (Ng et al., 2014; Lichtarge et al., 2019), factual error correction (Cao et al., 2020;…”
ReGAL: Refactoring Programs to Discover Generalizable Abstractions
Stengel-Eskin, E., Prasad, A., & Bansal, M. (2024) · arXiv · doi:10.48550/arxiv.2401.16467 · Cited by 1 · see citing works
PRD: Peer Rank and Discussion Improve Large Language Model based Evaluations
Li, R., Patel, T., & Du, X. (2023) · arXiv · doi:10.48550/arxiv.2307.02762
Woodpecker: Hallucination Correction for Multimodal Large Language Models
Yin, S., Fu, C., Zhao, S., et al. (2023) · arXiv · doi:10.48550/arxiv.2310.16045
Flagged first, then by citations
22 of the 152 references couldn’t be matched to a known paper.
Research works printed but not linked: 6
- Asai, 2024, Self-RAG: Learning to retrieve, generate, and critique through self-reflection
- Chen, 2023, Codet: Code generation with generated tests
- Chen, 2024, Universal self-consistency for large language models
- Jinlan, 2024, GPTScore: Evaluate as you desire, 6556
- Le, 2022, Coderl: Mastering code generation through pretrained models and deep reinforcement learning, 21314
- Valmeekam, 2023, Investigating the effectiveness of self-critiquing in LLMs solving planning tasks
Cited by
Sorted by citations
Large language models hallucination: A comprehensive survey
Alansari, A., & Luqman, H. (2026) · Computer Science Review · doi:10.1016/j.cosrev.2026.100970 · Cited by 120 · see citing works
Multi-Step Reasoning with Large Language Models, a Survey
Plaat, A., Wong, A., Verberne, S., Broekens, J., van Stein, B., & Bäck, T. (2025) · ACM Computing Surveys · doi:10.1145/3774896 · Cited by 131 · see citing works
“Section 4.1 provides guidance on how to choose reasoning approaches for different problem types.”
InfluentialToward LLM-agent-based modeling of transportation systems: A conceptual framework
Liu, T., Yang, J., & Yin, Y. (2025) · Artificial Intelligence for Transportation · doi:10.1016/j.ait.2025.100001 · Cited by 23 · see citing works
BioAgents: Bridging the gap in bioinformatics analysis with multi-agent systems
Mehandru, N., Hall, A. K., Melnichenko, O., Dubinina, Y., Tsirulnikov, D., Bamman, D., Alaa, A., Saponas, S., & Malladi, V. S. (2025) · Scientific Reports · doi:10.1038/s41598-025-25919-z · Cited by 19 · see citing works
“Several techniques have been proposed to enable a language model to correct its outputs based on internal evaluation, including: self-consistency 44 , self-correction 45,46 , self-evolution 47 , self-feedback 48 and self-evaluation 49 .”
A Review of Prompt Engineering Techniques for Large Language Models
Naser, M. Z. (2025) · International Journal of Human-Computer Interaction · doi:10.1080/10447318.2025.2607553 · Cited by 17 · see citing works
“Initial prompts establish an interpretive frame that influences how subsequent instructions are processed (similar to anchoring in cognitive psychology (Furnham & Boo, 2011)) (Kamoi et al., 2024).”
From Prompt to Drug: Toward Pharmaceutical Superintelligence
Zhavoronkov, A. A., Gennert, D. G., & Shi, J. (2026) · ACS Central Science · doi:10.1021/acscentsci.5c01473 · Cited by 10 · see citing works
Chung, D. J. H., Gao, Z., Kvasiuk, Y., Li, T., Münchmeyer, M., Rudolph, M., Sala, F., & Tadepalli, S. C. (2025) · Machine Learning Science and Technology · doi:10.1088/2632-2153/adfcb0 · Cited by 12 · see citing works
“In principle, given a large enough context window for prior thoughts and unlimited inference time, LLMs may be able to perform such very long thought processes, but currently available models (with a public reasoning chain) do not show very deep thought processes in our experience.”
Generative AI in Heritage Practice: Improving the Accessibility of Heritage Guidance
Witte, J., Lee, E., Brausem, L., Shillabeer, V., & Bonacchi, C. (2025) · Heritage · doi:10.3390/heritage8120513 · Cited by 10 · see citing works
Understanding Agentic AI: Algorithms and Infrastructure
Ma, W.-L., Guo, Y.-J., Han, Q.-L., Zhou, W., Zhu, X.-G., Xiong, J., Wen, S., & Xiang, Y. (2026) · IEEE/CAA Journal of Automatica Sinica · doi:10.1109/jas.2026.125993 · Cited by 11 · see citing works
GradeAgentOps: A Verification-First Framework for Evidence-Anchored LLM Exam Grading
Anghel, C., Anghel, A. A., Crăciun, M., Cocu, A., Vulpe, D. E., Andrei, C. A., Maier, C., Scheau, C., Dragosloveanu, Ş., & Cergan, R. (2026) · AI · doi:10.3390/ai7060198 · Cited by 8 · see citing works
“This motivates a controlled repair policy that is triggered by specific, machine-detectable violations rather than unconstrained regeneration [24].”
Confidence v.s. Critique: A Decomposition of Self-Correction Capability for LLMs
Yang, Z., Zhang, Y., Wang, Y., Xu, Z., Lin, J., & Sui, Z. (2025) · Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2025.acl-long.203 · Cited by 5 · see citing works
Self-citationHwang, W., Nguyen, T.-R.-G., & Shadiev, R. (2026) · Journal of Computer Assisted Learning · doi:10.1002/jcal.70208 · Cited by 7 · see citing works
Minerva: Evaluating Complex Video Reasoning
Nagrani, A., Menon, S., İşcen, A., Buch, S., Mehran, R., Jha, N., Hauth, A., Zhu, Y., Vondrick, C., Сиротенко, М., et al. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.02222 · Cited by 9 · see citing works
“Unsurprisingly, multiple works find that reference-based evaluation proves consistently more reliable [13], and that language models are poor evaluators of reasoning in isolation [22, 25, 44].”
Ibrahim, A., Alsukkar, M., Xu, Y., Hermaputi, R. L., Yin, S., & Xiao, Y. (2026) · Energy Sustainable Development/Energy for sustainable development · doi:10.1016/j.esd.2026.102104 · Cited by 10 · see citing works
Mitigating Context Bias in Vision–Language Models via Multimodal Emotion Recognition
Popescu, C.-B., Florea, L., & Florea, C. (2025) · Electronics · doi:10.3390/electronics14163311 · Cited by 4 · see citing works
“Additionally, self-feedback mechanisms have been shown to improve semantic grounding the alignment between image regions and generated text highlighting a promising direction for mitigating hallucinations and misinterpretations [11].”
Artificial Creativity: from predictive AI to Generative System 3
Chávez-Autor, J. C. (2025) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2025.1654716 · Cited by 3 · see citing works
“A recent survey maps when such loops help or fail across tasks (Kamoi et al., 2024), and domain-specific controllers demonstrate gains in code generation with explicit revise-and-retry cycles (Ding et al., 2024).”
InfluentialDetecting Semantic Clones of Unseen Functionality
Kitsios, K., Sovrano, F., Barr, E. T., & Bacchelli, A. (2025) · IEEE/ACM International Conference on Automated Software Engineering (ASE) · doi:10.1109/ase63991.2025.00112 · Cited by 3 · see citing works
“Through a system command, we required the LLM to generate the explanation prior to the decision, enabling the use of the explanation as a means of self-correction [74].”
Predictable artificial intelligence
Zhou, L., Casares, P. A. M., Martínez‐Plumed, F., Burden, J., Burnell, R., Cheke, L. G., Ferri, C. S., Marcoci, A., Mehrbakhsh, B., Moros-Daval, Y., et al. (2026) · Artificial Intelligence · doi:10.1016/j.artint.2026.104491 · Cited by 18 · 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
“Some approaches are validated only on small, specific constraint sets, limiting their general applicability (Schick et al., 2023; Saunders et al., 2022; Lee et al., 2024; Mu et al., 2023; Yang et al., 2022; Gou et al., 2024; Varshney et al., 2023).”
InfluentialDo NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs
Chen, X., Xu, J., Liang, T., He, Z., Pang, J., Yu, D., Song, L., Liu, Q., Zhou, M., Zhang, Z., et al. (2024) · arXiv · doi:10.48550/arxiv.2412.21187 · Cited by 3 · see citing works
Zhao, H., Zhang, L., Lian, K., Sun, F., Chen, B., Liu, Y., Wu, Z., Zhang, Y., & Yang, M. (2025) · IEEE/ACM International Conference on Automated Software Engineering (ASE) · doi:10.1109/ase63991.2025.00082 · Cited by 2 · see citing works
“This method is training-free and model-specific, and avoids problems in traditional self-correction methods like relying on feedback from external tools [33], [34].”
Self-citationImproving Next Tokens via Second-to-Last Predictions with Generate and Refine
Schneider, J. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-91398-3_17 · Cited by 1 · see citing works
“In particular, on a larger scale, our work contributes to the ongoing debate on when LLMs can self-correct during inference, where self-correction might also be triggered by external sources [9].”
InfluentialLLaVA-SCo: Teach Vision Language Models to Self-Correct
Liu, Z., Jiang, G., & Khajavi, S. H. (2025) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw67362.2025.00325 · Cited by 1 · see citing works
“We operate within the intrinsic self-correction framework [9, 10], where models attempt to rectify their initial responses without relying on external feedback.”
EMLC: An extensible multi-level correction framework for text-to-SQL
Lei, J., Tan, Y., & Wang, Y. (2026) · Information Processing & Management · doi:10.1016/j.ipm.2025.104560 · Cited by 1 · see citing works
Revisiting data analysis with Pre-trained foundation models
Liang, C., Yang, D., Liang, Z., Liang, Z., Zhang, T., Xiao, B., Yang, Y., Wang, W., & Wang, H. (2026) · The VLDB Journal · doi:10.1007/s00778-025-00953-5 · Cited by 1 · see citing works
Beyond Accuracy: The Role of Calibration in Self-Improving Large Language Models
Huang, L., Li, D. B., Liu, H., & Cheng, L. (2025) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata66926.2025.11400862 · Cited by 1 · see citing works
“Broadly, approaches to LLM self-improvement fall into two categories (25): intrinsic methods, where LLMs generate feedback on their own responses and attempt self-correction (6; 26), and external methods, where auxiliary tools such as code executors (27), search engines (28), or human feedback (29)…”
Adaptive Hierarchical Evaluation of LLMs and SAST tools for CWE Prediction in Python
Adnan, M., & Kuhn, C. C. N. (2026) · International Conference on Advances in Artificial Intelligence and Machine Learning (AAIML) · doi:10.1109/aaiml67890.2026.11498216 · Cited by 1 · see citing works
“Effective feedback mechanisms require precise vulnerability identification, specifically the CWE type, location, and nature of the flaw [5].”
InfluentialReformulate, Retrieve, Localize: Agents for Repository-Level Bug Localization
Caumartin, G., & Melo, G. (2026) · doi:10.1145/3786161.3788460 · Cited by 1 · see citing works
Sundaram, G., Venktesh, B., V, & Sundharakumar, K. B. (2026) · Information Processing & Management · doi:10.1016/j.ipm.2026.105031 · Cited by 1 · see citing works
Naser, M. Z. (2026) · Language Resources and Evaluation · doi:10.1007/s10579-026-09938-4 · Cited by 2 · see citing works
Mendoza, D. E., Mavridou, A., Katis, A., et al. (2026) · IEEE/ACM International Conference on Software Engineering (ICSE) · doi:10.1145/3744916.3787815 · Cited by 1 · see citing works
Necessary and sufficient knowledge enhanced collaborative logical reasoning in LLMs
Wang, P., Ding, X., Xiong, K., et al. (2025) · Neural Networks · doi:10.1016/j.neunet.2025.108164 · Cited by 1 · see citing works
When AI Reviews Itself: Zero Answer Changes Across 72 Self-Correction RoundsPreprint
Hamidavi, A. (2026) · Research Square · doi:10.21203/rs.3.rs-10463884/v1 · Cited by 1 · see citing works
Talk Isn't Always Cheap: Understanding Failure Modes in Multi-Agent Debate
Wynn, A., Satija, H., & Hadfield, G. (2025) · arXiv · doi:10.48550/arxiv.2509.05396 · Cited by 1 · see citing works
“Self-refinement involves iteratively revising or adapting a model’s responses based on previous outputs, prompting the model to intentionally reflect on its existing responses and correct any mistakes (Kamoi et al., 2024).”
The Illusion of Diminishing Returns: Measuring Long Horizon Execution in LLMs
Sinha, A., Arun, A., Goel, S., Staab, S., & Geiping, J. (2025) · arXiv · doi:10.48550/arxiv.2509.09677 · Cited by 1 · see citing works
“As a training-free fix for our setting, we also try model-based self-verification (see Kamoi et al. (2024) for a survey) at each turn in Appendix C.1, which can be considered a form of self-refinement (Madaan et al., 2023) without having access to environment feedback (Gou et al., 2023).”
Vladika, J., Domres, A., Nguyen, M., Moser, R., Nano, J., Busch, F., Adams, L. C., Bressem, K. K., Bernhardt, D., Combs, S. E., et al. (2025) · arXiv · doi:10.48550/arxiv.2505.24830 · Cited by 1 · see citing works
Tsui, K. (2025) · arXiv · doi:10.48550/arxiv.2507.02778
“Recent work has highlighted intrinsic self-correction, where LLMs generate feedback on their own outputs (Shinn et al., 2023; Madaan et al., 2023; Kim et al., 2023; Kamoi et al., 2024a), as a way to improve performance, but critical limitations persist.”
Agentic Correlates of Consciousness and the Pursuit of Artificial General Intelligence
Gray, S., Tambouratzis, G., Venugopal, S., Raghavan, S., Tong, R. J., & Han, Z. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-00686-8_21
Title Generative Ai as a Fuzzy Validation Partner: A Perspective on Methodology and JudgmentPreprint
van Capelleveen, J., van de Wetering, R., Alechina, N., Janssens, G., & Wenniger, G. M. D. B. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5402161
AoPSFT: Subplan-Aware Fine-Tuning for Interpretable Multi-turn Reasoning Agents
Zhou, X., Geng, J., Cai, C., Li, L., & Zeng, D. D. (2025) · Communications in computer and information science · doi:10.1007/978-981-95-4088-4_12
Memory-Augmented Agent Training for Business Document Understanding
Liu, J., Zeng, Y., Højmark‐Bertelsen, M., Gadeberg, M. N., Wang, H., & Wu, Q. (2026) · Communications in computer and information science · doi:10.1007/978-3-032-11477-8_5
Fixing Broken Graphs: LLM-Powered Automatic Code Optimization for DNN Programs
Wang, H.-T., Sui, Y.-C., Xie, Y., Liu, Y., Sun, Y., Shi, C.-Q., & Zhang, Y. (2025) · International Conference on Automated Software Engineering · doi:10.1109/ase63991.2025.00144
“This design enables robust handling of diverse fragmentation patterns while maintaining reliability through systematic validation and self-correction capabilities [7].”
Self-citationDevin, A. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.6135630
Architectural Laws And Derived Constraints Of Verification And Consequential AI ReliancePreprint
Devin, A. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.6135727
Fireseed: A Prompt-Driven Self-Terminating Personality Module for Large Language Models
Pang, C., Yan, B., Zhu, Y., et al. (2025) · doi:10.1109/iccbdai66607.2025.11388826
Khatun, M. E., Noureddine, L., Sui, Z., & Ali-Gombe, A. (2025) · Annual Computer Security Applications Conference · doi:10.1109/acsac67867.2025.00021 · Cited by 1 · see citing works
Ren, Y., Zhou, L., Zhang, S., et al. (2025) · doi:10.1109/pcce68336.2025.11407205
“These robust estimators can be categorized based on the convexity of their influence functions: the Cauchy [46], Correction [47], Geman-McClure [48], Welsch [49], and Lorentzian [50] estimators employ non-convex functions, whereas the Fair [51], Kong [52], and QWLS [53] estimators utilize convex…”
Guiding LLM-Based Loop Invariant Synthesis via Feedback on Local Reasoning Errors
Li, T., Yan, Z., Liu, J., Di, P., & Zhang, X. (2026) · ACM Transactions on Programming Languages and Systems · doi:10.1145/3806652
“A key capability of modern LLMs is their ability to refine their outputs based on feedback [21].”
ReflectCAP: Detailed Image Captioning with Reflective Memory
Min, K., Kim, M., Lee, K.-I., et al. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37556-8_27
Zhong, M., Shi, Z., Wang, Z., Liu, R., Fang, M., & Chen, L. (2026) · Lecture notes in computer science · doi:10.1007/978-981-95-7078-2_37
SVCF: Enabling Zero-Shot Correction of Reasoning Steps in Multi-Modal Large Language Models
Jiang, B.-Y., Huang, T.-X., Yang, Y., Zhang, Y., Duan, G., & He, T. (2026) · IEEE International Conference on Acoustics, Speech, and Signal Processing · doi:10.1109/icassp55912.2026.11462050
Self-citationHeo, S. (2026) · Scientific Reports · doi:10.1038/s41598-026-50127-8 · Cited by 2 · see citing works
“11, 12, 13, the pipeline was evaluated on 45 sessions of the Korean Professional Engineer Structural Engineering examination (1350 problems in total).”
What Structural Intelligence Is Intelligence as Answerable Revision Under ConstraintPreprint
Jovanovic, V. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.6683258
Benchmarking and enhancing the ability of large language models on event generalization
Shao, S., Wen, T., Dong, Y., et al. (2026) · Information Processing & Management · doi:10.1016/j.ipm.2026.104851
Zhang, G., Li, W., Ye, X., et al. (2026) · Communications in computer and information science · doi:10.1007/978-981-95-7254-0_17
Esina, S. V., & Inkpen, D. Z. (2026) · Communications in computer and information science · doi:10.1007/978-3-032-20322-9_27
Learned Root-Cause SQL Prioritization and Diagnosis for Complex Database Performance Issues
Luo, K., Li, S., Shen, M., et al. (2026) · Proceedings - ACM-SIGMOD International Conference on Management of Data · doi:10.1145/3788853.3803087
“To ensure both correctness and utility, we design an iterative Generation–Review dual-agent workflow inspired by self-correction techniques [20, 36](see Figure 5).”
Iterative Cross-LLM Agent Framework for Personalized Resume Optimization
Barathi, B. U. A., Bhuvanesh, R., & Dakshinaa, K. S. (2026) · doi:10.1109/ic-icns68863.2026.11537727
“Initially this was a great idea but when used, it lead to monotony bias [14], since the model does not question its own assumptions, leading to stagnation rather than improvement.”
Anomalies in AI Outputs Beyond Input Data Quality: The Significance of Reasoning
Royo, J. [. ]. A., Martín, Ó., Martínez, E., Montalvo, F. J. F., & Blanco‐Fernández, J. (2026) · Applied Sciences · doi:10.3390/app16115491
Feynman Explanation Prompting: A Pedagogically Grounded Framework for Large Language Models
Bildirici, F., Karakuş, M., & Kızıltepe, R. S. (2026) · International Congress on Human-Computer Interaction, Optimization and Robotic Applications (ICHORA) · doi:10.1109/ichora69329.2026.11537090 · Cited by 1 · see citing works
“…contextual coherence and reflective depth [9], [11]; self-correction research reveals both the promise and the boundaries of model-driven revision [13]–[15]; the Feynman method strengthens teachability through iterative explanation and re-teaching [1], [2], [16]; Feynman-inspired frameworks…”
Multi-Agent LLM Classification of STPA Context Tables
Katranas, I., Dokas, Ι. M., & Kafoutis, G. C. E. (2026) · New Trends in Civil Aviation (NTCA) · doi:10.23919/ntca68808.2026.11524157
“This approach allows the LLM to examine its initial classification and refine it based on the system safety requirements [27], [28].”
Case-Based Adaptation and Retrieval-Augmented Self-reflection for Multi-label Classification
Jayawardena, L., Wiratunga, N., Nkisi-Orji, I., & Nicol, D. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-33865-5_5
O., E. F. T. (2026) · Research Square · doi:10.21203/rs.3.rs-10084276/v1
Reflect: Transparent Principle-Guided Reasoning for Constitutional Alignment at Scale
Bell, H., Zhang, C., Haque, M. M., Zaman, S., Potdar, D., & Fain, B. (2026) · ACM Conference on Fairness, Accountability, and Transparency (FAccT) · doi:10.1145/3805689.3806454
“…iterative self-refinement can be effective at improving task-specific performance including math reasoning, sentiment reversal, and dialogue generation [13, 56, 65, 78], and at reducing hallucinations [31], though other work has questioned the effectiveness of such self-correction [33, 40, 52].”
Kostka, A., & Chudziak, J. A. (2026) · Knowledge · doi:10.3390/knowledge6030017
“Indeed, without independent evaluation, self-level refinements can even degrade an otherwise accurate response [44,45].”
RILEE: Reasoning-Intensive Legal Element Extraction via Hierarchical Re-answering
Wang, L., Hu, J., Ying, Y., Liu, Z., Zhou, Q., Guo, D., & Luo, C. (2026) · Lecture notes in computer science · doi:10.1007/978-981-92-2759-4_4
Post-Generation Audit for Detecting Unsupported Claims in KPI-Grounded LLM Performance Feedback
Belarache, O., Hannouni, S., Ounacer, S., et al. (2026) · International Conference on Intelligent Systems and Digital Applications (ISDA) · doi:10.1109/isda70544.2026.11606099
“Kamoi et al. survey self-correction methods and conclude that there is still no consensus on when LLMs can actually correct their own mistakes [5].”
Localized Diffusion Repair for Reliable Tool-Using LLM Agents
Wang, X., Wei, F., Yang, S., Li, Y., Wu, J. H., & Chen, J. (2026) · International Conference on Algorithms, Software Engineering and Network Security (ASENS) · doi:10.1109/asens69964.2026.11605180
“Existing systems mitigate these issues with constrained decoding, schema-aware prompting, validation-and-retry loops, or self-refinement that regenerates the full call [15], [16].”
From Evaluation to Optimisation: Hierarchy-Aware Training Signals for CWE Prediction in Python
Adnan, M., Srun, M., & Kuhn, C. C. N. (2026) · Machine Learning and Knowledge Extraction · doi:10.3390/make8090258
Son, B., Jung, M., & Lee, S. (2026) · The Journal of Korean Institute of Information Technology · doi:10.14801/jkiit.2026.24.7.81
AI That Fixes Its Own Mistakes: A Self-healing Agentic Model Without Retraining
Singh, A., Chanda, P., Basak, R., et al. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-28591-1_24
AI That Fixes Its Own Mistakes: A Self-Healing Agentic Model without Retraining
Singh, A., Chanda, P., Basak, R., et al. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-28310-8_9
Reasoning over Semantic IDs Enhances Generative Recommendation
He, Y., Sun, Y., Tan, J., Chen, Y., Kong, X., Shen, C., Wang, X., Zhang, A., & Chua, T. (2026) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3770855.3818122 · Cited by 1 · see citing works
“The quality of reasoning can be further improved by allocating additional inference-time computation, for example through parallel sampling [36, 51, 58], and iterative refinement [26, 28, 55].”
Hua, L., Zheng, J., Zheng, H., et al. (2026) · doi:10.1109/ccssta69471.2026.11635782
“…self-consistency [17] yields no real gain under a controlled DSL (oracle@5 exceeds single-candidate by only 2.2 pp; average distinct result clusters per set is 1.12), because accuracy gaps stem from systematic DSL expressiveness ceilings and domain-term mapping gaps, consistent with [8], [9].”
InfluentialReassessing One-Round Test-Time Refinement for Code Generation
Kim, J. (2026) · AI · doi:10.3390/ai7090353
Cognitive alignment: bridging EEG and vision via cascaded semantic reasoning
Huang, Y., Weng, Z., Zheng, Z., et al. (2026) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2026.111320
Gong, Y., Shao, M., Lian, X., & Zhou, J. (2026) · Applied Sciences · doi:10.3390/app16178462
Liu, J., Wang, C., & Li, H. (2026) · Symmetry · doi:10.3390/sym18091437
Stephen, S., & Vignesh, R. (2026) · Computers & Electrical Engineering · doi:10.1016/j.compeleceng.2026.111514
Balancing Accuracy and Confidence in LLMs: Reinforcement Learning with Entropy-Calibrated Rewards
Kato, K., Chu, C., Murawaki, Y., & Kurohashi, S. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-32732-1_12
When Iterative Prompting Fails: An Empirical Study of Unit Test Generation with Open-Source LLMs
Tran, M.-D., & Mai, K. (2026) · International Conference on Multimedia Analysis and Pattern Recognition (MAPR) · doi:10.1109/mapr72750.2026.11685714
“More recent studies revisit the reliability of intrinsic self-correction under prompt-only settings [11, 12], suggesting that such approaches remain unstable, particularly for code-centric tasks requiring precise semantic alignment.”
iBRAHIMOĞLU, N., Demi̇rezen, M. U., & Yıldız, F. (2026) · Cluster Computing · doi:10.1007/s10586-026-06537-4
Vladika, J., Domres, A., Nguyen, M., Moser, R., Nano, J., Busch, F., Adams, L. C., Bressem, K. K., Bernhardt, D., Combs, S. E., et al. (2026) · Journal of Medical Internet Research · doi:10.2196/92090 · Cited by 4 · see citing works
“Hence, post hoc methods are emerging, where LLMs self-correct their responses only after they are generated [8].”
Analysis of architecture options for foundation-model-based agents: A taxonomy and decision model
Zhou, J., Lu, Q., Chen, J., Zhu, L., Xu, X., Xing, Z., & Harrer, S. (2026) · Journal of Systems and Software · doi:10.1016/j.jss.2026.113109 · Cited by 2 · see citing works
Agentic Semantic Communication: Concepts, Architectures, and Visions
Seon, J., Lee, S., Kim, J. Y., Sun, Y. G., Seo, H., Bennis, M., Kim, D. I., & Kim, J. Y. (2026) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2026.3735868
“4) Self-Correction: In LLMs, the ability to autonomously identify and correct errors is referred to as self-correction [142].”
Enhancing Autonomous Agent Accuracy Through Reflection-Based Reasoning
Pang, L., Cen, C., Yu, J., et al. (2026) · doi:10.1109/apan70967.2026.11709799
“Recent survey work has also examined the capabilities and limitations of LLM self-correction across a broad range of tasks and settings [19].”
ETTAFSSAOUI, Y., & Bencharef, O. (2026) · Research Square · doi:10.21203/rs.3.rs-10963932/v1
Can LLMs Self-Correct Table Reasoning Errors?
Shaikh, F. (2026) · Proceedings of the First Workshop on Structured Understanding, Retrieval, and Generation in the LLM Era (SURGeLLM 2026) · doi:10.18653/v1/2026.surgellm-1.19
“Kamoi et al. (2024), in a critical survey published in TACL, concluded that “no prior work demonstrates successful self-correction with feedback from prompted LLMs, except for studies in tasks that are exceptionally suited for self-correction.””
Weatherhead, S., Salim, F., & Belbasis, A. (2025) · arXiv · doi:10.48550/arxiv.2510.18254
“Surveys and targeted studies, however, suggest that many gains attributed to reflection depend on external signals—tests, retrieval, or explicit error flags—with limited success at autonomous corrective reasoning when those signals are weak[3, 12–14].”
PatchWorld: Gradient-Free Optimization of Executable World Models for Agent Environments
Bai, J., Guo, Y., Dong, Y., Xiong, J., Zheng, T., Li, Y., Fang, T., Li, Y., Gao, Y., Huang, H., et al. (2026) · arXiv · doi:10.48550/arxiv.2605.30880
Can LLMs Correct Themselves? A Benchmark of Self-Correction in LLMs
Tie, G., Yuan, Z., Zhao, Z., Hu, C., Gu, T., Zhang, R., Zhang, S., Wu, J., Tu, X., Jin, M., et al. (2025) · arXiv · doi:10.48550/arxiv.2510.16062
“Building on a recent survey of self-correction approaches [29], we categorize such methods into three primary categories (i.e., intrinsic correction , external correction and fine-tuned correction ), and select 11 representative methods from them.”
Self-citation
- Authors
- Kamoi, R., Zhang, Y., Zhang, N., Han, J., & Zhang, R.
- Affiliations
- Pennsylvania State University (US)
- University of Illinois Urbana-Champaign (US)
- Venue
- Transactions of the Association for Computational Linguistics
- Publisher
- MIT Press
- Published
- 2024
- Volume
- 12
- Pages
- 1417-1440
- ISSN
- 2307-387X (ISSN-L)
- OpenAlex
- W4404131920
- Semantic Scholar
- 3fa1e1c67514b9eaf9ec8da562baef8974b4f3f9
- Cited by
- 82
- References
- 152
- Citations by year
- 2026: 59 · 2025: 23
- FWCI
- 13.10
- Type
- journal article
- Language
- en
- Open access
- diamond
- doi.org
- Locations
- Transactions of the Association for Computational Linguistics (publishedVersion, cc-by)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
- Physical Sciences › Computer Science › Artificial Intelligence › Speech and dialogue systems
- Keywords
- self-correction, large language models, self-assessment, external feedback, assessment fairness
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
