{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,20]],"date-time":"2026-08-20T14:43:06Z","timestamp":1787236986174,"version":"build-2736575974"},"reference-count":156,"publisher":"MIT Press","license":[{"start":{"date-parts":[[2024,5,6]],"date-time":"2024-05-06T00:00:00Z","timestamp":1714953600000},"content-version":"vor","delay-in-days":126,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["direct.mit.edu"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2024,5,3]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>While large language models (LLMs) have shown remarkable effectiveness in various NLP tasks, they are still prone to issues such as hallucination, unfaithful reasoning, and toxicity. A promising approach to rectify these flaws is correcting LLMs with feedback, where the LLM itself is prompted or guided with feedback to fix problems in its own output. Techniques leveraging automated feedback\u2014either produced by the LLM itself (self-correction) or some external system\u2014are of particular interest as they make LLM-based solutions more practical and deployable with minimal human intervention. This paper provides an exhaustive review of the recent advances in correcting LLMs with automated feedback, categorizing them into training-time, generation-time, and post-hoc approaches. We also identify potential challenges and future directions in this emerging field.<\/jats:p>","DOI":"10.1162\/tacl_a_00660","type":"journal-article","created":{"date-parts":[[2024,5,6]],"date-time":"2024-05-06T16:13:22Z","timestamp":1715012002000},"page":"484-506","update-policy":"https:\/\/doi.org\/10.1162\/mitpressjournals.corrections.policy","source":"Crossref","is-referenced-by-count":82,"title":["Automatically Correcting Large Language Models:\n                    <i>Surveying the Landscape of Diverse Automated Correction Strategies<\/i>"],"prefix":"10.1162","volume":"12","author":[{"given":"Liangming","family":"Pan","sequence":"first","affiliation":[{"name":"University of California, Santa Barbara, USA. liangmingpan@ucsb.edu"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Michael","family":"Saxon","sequence":"additional","affiliation":[{"name":"University of California, Santa Barbara, USA. saxon@ucsb.edu"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wenda","family":"Xu","sequence":"additional","affiliation":[{"name":"University of California, Santa Barbara, USA. wendaxu@ucsb.edu"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Deepak","family":"Nathani","sequence":"additional","affiliation":[{"name":"University of California, Santa Barbara, USA. dnathani@ucsb.edu"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xinyi","family":"Wang","sequence":"additional","affiliation":[{"name":"University of California, Santa Barbara, USA. xinyi_wang@ucsb.edu"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"William Yang","family":"Wang","sequence":"additional","affiliation":[{"name":"University of California, Santa Barbara, USA. william@cs.ucsb.edu"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"281","published-online":{"date-parts":[[2024,5,3]]},"reference":[{"key":"2024050620131470100_bib1","doi-asserted-by":"publisher","first-page":"7716","DOI":"10.18653\/v1\/2023.acl-long.427","article-title":"RL4F: Generating natural language feedback with reinforcement learning for repairing model outputs","volume-title":"Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (ACL)","author":"Aky\u00fcrek","year":"2023"},{"key":"2024050620131470100_bib2","doi-asserted-by":"publisher","first-page":"25","DOI":"10.3115\/v1\/E14-2007","article-title":"CASMACAT: A computer-assisted translation workbench","volume-title":"Proceedings of the 14th Conference of the European Chapter of the Association for Computational Linguistics (EACL)","author":"Alabau","year":"2014"},{"key":"2024050620131470100_bib3","article-title":"Training a helpful and harmless assistant with reinforcement learning from human feedback","author":"Bai","year":"2022","journal-title":"CoRR"},{"key":"2024050620131470100_bib4","article-title":"Constitutional AI: harmlessness from AI feedback","author":"Bai","year":"2022","journal-title":"CoRR"},{"key":"2024050620131470100_bib5","article-title":"Large linguistic models: Analyzing theoretical linguistic abilities of LLMs","author":"Begus","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib6","doi-asserted-by":"publisher","first-page":"1125873","DOI":"10.3389\/fpsyg.2023.1125873","article-title":"Daily automated feedback enhances self-regulated learning: A longitudinal randomized field experiment","volume":"14","author":"Bellh\u00e4user","year":"2023","journal-title":"Frontiers in Psychology"},{"key":"2024050620131470100_bib7","doi-asserted-by":"publisher","first-page":"3659","DOI":"10.18653\/v1\/D19-1378","article-title":"Global reasoning over database structures for text-to-SQL parsing","volume-title":"Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)","author":"Bogin","year":"2019"},{"issue":"2","key":"2024050620131470100_bib8","doi-asserted-by":"publisher","first-page":"99","DOI":"10.1177\/0022167883232011","article-title":"Reflective learning: Key to learning from experience","volume":"23","author":"Boyd","year":"1983","journal-title":"Journal of Humanistic Psychology"},{"key":"2024050620131470100_bib9","doi-asserted-by":"publisher","first-page":"6251","DOI":"10.18653\/v1\/2020.emnlp-main.506","article-title":"Factual error correction for abstractive summarization models","volume-title":"Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)","author":"Cao","year":"2020"},{"key":"2024050620131470100_bib10","doi-asserted-by":"publisher","first-page":"6491","DOI":"10.18653\/v1\/2021.emnlp-main.522","article-title":"Editing factual knowledge in language models","volume-title":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing (EMNLP)","author":"De Cao","year":"2021"},{"key":"2024050620131470100_bib11","article-title":"A new era in software security: Towards self-healing software via large language models and formal verification","author":"Charalambous","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib12","article-title":"Improving code generation by training with natural language feedback","author":"Chen","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib13","article-title":"Codet: Code generation with generated tests","volume-title":"Proceedings of the 11th International Conference on Learning Representations (ICLR)","author":"Chen","year":"2023"},{"key":"2024050620131470100_bib14","article-title":"Reconcile: Round-table conference improves reasoning via consensus among diverse LLMs","author":"Chen","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib15","article-title":"Iterative translation refinement with large language models","author":"Chen","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib16","article-title":"Teaching large language models to self-debug","author":"Chen","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib17","article-title":"Factool: Factuality detection in generative AI \u2013 a tool augmented framework for multi-task and multi-domain scenarios","author":"Chern","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib18","doi-asserted-by":"publisher","first-page":"7282","DOI":"10.18653\/v1\/2021.acl-long.565","article-title":"All that\u2019s \u2018human\u2019 is not gold: Evaluating human evaluation of generated text","volume-title":"Processings of the 59th Annual Meeting of the Association for Computational Linguistics (ACL)","author":"Clark","year":"2021"},{"key":"2024050620131470100_bib19","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-main.778","article-title":"LM vs LM: Detecting factual errors via cross examination","author":"Cohen","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib20","article-title":"Faithful reasoning using large language models","author":"Creswell","year":"2022","journal-title":"CoRR"},{"key":"2024050620131470100_bib21","article-title":"Language models show human-like content effects on reasoning","author":"Dasgupta","year":"2022","journal-title":"CoRR"},{"key":"2024050620131470100_bib22","article-title":"Plug and play language models: A simple approach to controlled text generation","volume-title":"Proceedings of the 8th International Conference on Learning Representations (ICLR)","author":"Dathathri","year":"2020"},{"issue":"2","key":"2024050620131470100_bib23","doi-asserted-by":"publisher","first-page":"101","DOI":"10.1007\/s10590-020-09252-y","article-title":"A review of the state-of-the-art in automatic post-editing","volume":"35","author":"do Carmo","year":"2021","journal-title":"Machine Translation"},{"key":"2024050620131470100_bib24","article-title":"Improving factuality and reasoning in language models through multiagent debate","author":"Yilun","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib25","article-title":"Alpacafarm: A simulation framework for methods that learn from human feedback","author":"Dubois","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib26","doi-asserted-by":"publisher","first-page":"2214","DOI":"10.18653\/v1\/P19-1213","article-title":"Ranking generated summaries by correctness: An interesting but challenging application for natural language inference","volume-title":"Proceedings of the 57st Annual Meeting of the Association for Computational Linguistics (ACL)","author":"Falke","year":"2019"},{"key":"2024050620131470100_bib27","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00626","article-title":"Bridging the gap: A survey on integrating (human) feedback for natural language generation","author":"Fernandes","year":"2023","journal-title":"CoRR"},{"issue":"3","key":"2024050620131470100_bib28","doi-asserted-by":"publisher","first-page":"156","DOI":"10.1093\/pch\/pxy102","article-title":"Catch the moment: The power of turning mistakes into \u2018precious\u2019 learning opportunities","volume":"24","author":"Ferretti","year":"2019","journal-title":"Paediatrics & Child Health"},{"key":"2024050620131470100_bib29","doi-asserted-by":"publisher","DOI":"10.1145\/3611643.3616243","article-title":"Baldur: Whole-proof generation and repair with large language models","author":"First","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib30","doi-asserted-by":"publisher","first-page":"811","DOI":"10.1162\/tacl_a_00491","article-title":"High quality rather than high model probability: Minimum bayes risk decoding with neural metrics","author":"Freitag","year":"2022","journal-title":"Transactions of the Association for Computational Linguistics (TACL)"},{"key":"2024050620131470100_bib31","article-title":"Improving language model negotiation with self-play and in-context learning from AI feedback","author":"Yao","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib32","article-title":"The capacity for moral self-correction in large language models","author":"Ganguli","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib33","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-main.27","article-title":"Continually improving extractive QA via human feedback","author":"Ge","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib34","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.acl-long.910","article-title":"Rarr: Researching and revising what language models say, using language models","volume-title":"Proceedings of the 61th Annual Meeting of the Association for Computational Linguistics (ACL)","author":"Gao","year":"2023"},{"key":"2024050620131470100_bib35","doi-asserted-by":"publisher","first-page":"3356","DOI":"10.18653\/v1\/2020.findings-emnlp.301","article-title":"RealToxicityPrompts: Evaluating neural toxic degeneration in language models","volume-title":"Findings of the Association for Computational Linguistics: EMNLP 2020","author":"Gehman","year":"2020"},{"key":"2024050620131470100_bib36","article-title":"Self-verification improves few-shot clinical information extraction","author":"Gero","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib37","article-title":"Improving alignment of dialogue agents via targeted human judgements","author":"Glaese","year":"2022","journal-title":"CoRR"},{"key":"2024050620131470100_bib38","article-title":"Aligning language models with preferences through f-divergence minimization","author":"Go","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib39","article-title":"ROSCOE: A suite of metrics for scoring step-by-step reasoning","volume-title":"Proceedings of the 11th International Conference on Learning Representations (ICLR)","author":"Golovneva","year":"2023"},{"key":"2024050620131470100_bib40","article-title":"CRITIC: Large language models can self-correct with tool-interactive critiquing","author":"Gou","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib41","article-title":"Reinforced self- training (rest) for language modeling","author":"Gulcehre","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib42","article-title":"How close is chatgpt to human experts? Comparison corpus, evaluation, and detection","author":"Guo","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib43","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-main.507","article-title":"Reasoning with language model is planning with world model","author":"Hao","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib44","article-title":"Rethinking with retrieval: Faithful large language model inference","author":"He","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib45","article-title":"Deberta: Decoding-enhanced bert with disentangled attention","volume-title":"Proceedings of The 9th International Conference on Learning Representations (ICLR)","author":"He","year":"2021"},{"key":"2024050620131470100_bib46","article-title":"LLM self defense: By self examination, LLMs know they are being tricked","author":"Helbling","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib47","doi-asserted-by":"publisher","first-page":"11548","DOI":"10.18653\/v1\/2023.findings-acl.733","article-title":"Detecting edit failures in large language models: An improved specificity benchmark","volume-title":"Findings of the Association for Computational Linguistics: ACL 2023","author":"Hoelscher-Obermaier","year":"2023"},{"key":"2024050620131470100_bib48","doi-asserted-by":"publisher","first-page":"1638","DOI":"10.18653\/v1\/P18-1152","article-title":"Learning to write with cooperative discriminators","volume-title":"Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (ACL)","author":"Holtzman","year":"2018"},{"key":"2024050620131470100_bib49","article-title":"A closer look at the self-verification abilities of large language models in logical reasoning","author":"Hong","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib50","article-title":"Large language models cannot self-correct reasoning yet","author":"Huang","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib51","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-main.67","article-title":"Large language models can self-improve","author":"Huang","year":"2022","journal-title":"CoRR"},{"key":"2024050620131470100_bib52","article-title":"Selfevolve: A code evolution framework via large language models","author":"Jiang","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib53","doi-asserted-by":"publisher","first-page":"1266","DOI":"10.18653\/v1\/2022.emnlp-main.82","article-title":"Maieutic prompting: Logically consistent reasoning with recursive explanations","volume-title":"Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing (EMNLP)","author":"Jung","year":"2022"},{"key":"2024050620131470100_bib54","article-title":"Language models (mostly) know what they know","author":"Kadavath","year":"2022","journal-title":"CoRR"},{"key":"2024050620131470100_bib55","article-title":"CritiqueLLM: Scaling LLM-as-critic for effective and explainable evaluation of large language model generation","author":"Ke","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib56","article-title":"Bertrand-dr: Improving text-to-sql using a discriminative re-ranker","author":"Kelkar","year":"2020","journal-title":"CoRR"},{"key":"2024050620131470100_bib57","article-title":"Discriminator-guided multi-step reasoning with language models","author":"Khalifa","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib58","article-title":"Language models can solve computer tasks","author":"Kim","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib59","article-title":"Overcoming catastrophic forgetting in neural networks","author":"Kirkpatrick","year":"2016","journal-title":"CoRR"},{"key":"2024050620131470100_bib60","article-title":"Large language models are zero-shot reasoners","volume-title":"Proceedings of the 2022 Annual Conference on Neural Information Processing Systems (NeurIPS)","author":"Kojima","year":"2022"},{"key":"2024050620131470100_bib61","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/N18-3012","article-title":"Can neural machine translation be improved with user feedback?","volume-title":"Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HIT)","author":"Kreutzer","year":"2018"},{"key":"2024050620131470100_bib62","article-title":"Coderl: Mastering code generation through pretrained models and deep reinforcement learning","volume-title":"Proceedings of the Annual Conference on Neural Information Processing Systems (NeurIPS)","author":"Le","year":"2022"},{"key":"2024050620131470100_bib63","doi-asserted-by":"publisher","first-page":"6045","DOI":"10.18653\/v1\/D19-1624","article-title":"Clause-wise and recursive decoding for complex and cross-domain text-to- SQL generation","volume-title":"Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)","author":"Lee","year":"2019"},{"key":"2024050620131470100_bib64","doi-asserted-by":"publisher","first-page":"438","DOI":"10.18653\/v1\/2022.findings-acl.37","article-title":"Plug-and-play adaptation for continuously-updated QA","volume-title":"Findings of the Association for Computational Linguistics: ACL 2022","author":"Lee","year":"2022"},{"key":"2024050620131470100_bib65","doi-asserted-by":"publisher","first-page":"3685","DOI":"10.18653\/v1\/2021.eacl-main.322","article-title":"Adaptation of back-translation to automatic post-editing for synthetic data generation","volume-title":"Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics (EACL)","author":"Lee","year":"2021"},{"key":"2024050620131470100_bib66","doi-asserted-by":"publisher","first-page":"2407","DOI":"10.18653\/v1\/2022.emnlp-main.154","article-title":"SafeText: A benchmark for exploring physical safety in language models","volume-title":"Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing (EMNLP)","author":"Levy","year":"2022"},{"key":"2024050620131470100_bib67","doi-asserted-by":"publisher","first-page":"4718","DOI":"10.18653\/v1\/2021.findings-acl.416","article-title":"Investigating memorization of conspiracy theories in text generation","volume-title":"Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021","author":"Levy","year":"2021"},{"key":"2024050620131470100_bib68","article-title":"Halueval: A large-scale hallucination evaluation benchmark for large language models","author":"Li","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib69","article-title":"Self-checker: Plug-and-play modules for fact-checking with large language models","author":"Li","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib70","article-title":"PRD: Peer rank and discussion improve large language model based evaluations","author":"Li","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib71","article-title":"Diffusion-lm improves controllable text generation","volume-title":"Proceedings of the Annual Conference on Neural Information Processing Systems (NeurIPS)","author":"Li","year":"2022"},{"key":"2024050620131470100_bib72","doi-asserted-by":"publisher","first-page":"5315","DOI":"10.18653\/v1\/2023.acl-long.291","article-title":"Making language models better reasoners with step-aware verifier","volume-title":"Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (ACL)","author":"Li","year":"2023"},{"key":"2024050620131470100_bib73","article-title":"Let\u2019s verify step by step","author":"Lightman","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib74","doi-asserted-by":"publisher","first-page":"3214","DOI":"10.18653\/v1\/2022.acl-long.229","article-title":"TruthfulQA: Measuring how models mimic human falsehoods","volume-title":"Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (ACL)","author":"Lin","year":"2022"},{"key":"2024050620131470100_bib75","article-title":"LLM-eval: Unified multi-dimensional automatic evaluation for open-domain conversations with large language models","author":"Lin","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib76","article-title":"Generating with confidence: Uncertainty quantification for black-box large language models","author":"Lin","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib77","article-title":"Chain of hindsight aligns language models with feedback","author":"Liu","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib78","doi-asserted-by":"publisher","first-page":"11557","DOI":"10.18653\/v1\/2023.emnlp-main.708","article-title":"Crystal: Introspective reasoners reinforced with self-feedback","volume-title":"Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing (EMNLP)","author":"Liu","year":"2023"},{"key":"2024050620131470100_bib79","doi-asserted-by":"publisher","first-page":"1065","DOI":"10.18653\/v1\/2021.acl-short.135","article-title":"Simcls: A simple framework for contrastive learning of abstractive summarization","volume-title":"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (ACL\/IJCNLP)","author":"Liu","year":"2021"},{"key":"2024050620131470100_bib80","doi-asserted-by":"publisher","first-page":"261","DOI":"10.1146\/annurev-orgpsych-120920-044531","article-title":"Developing self-awareness: Learning processes for self-and interpersonal growth","volume":"10","author":"London","year":"2023","journal-title":"Annual Review of Organizational Psychology and Organizational Behavior"},{"key":"2024050620131470100_bib81","article-title":"QUARK: Controllable text generation with reinforced unlearning","volume-title":"Proceedings of the Annual Conference on Neural Information Processing Systems (NeurIPS)","author":"Ximing","year":"2022"},{"key":"2024050620131470100_bib82","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-main.1036","article-title":"New trends in machine translation using large language models: Case examples with chatgpt","author":"Lyu","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib83","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.ijcnlp-main.20","article-title":"Faithful chain-of-thought reasoning","author":"Lyu","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib84","doi-asserted-by":"publisher","first-page":"2833","DOI":"10.18653\/v1\/2022.emnlp-main.183","article-title":"Memory-assisted prompt editing to improve GPT-3 after deployment","volume-title":"Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing (EMNLP)","author":"Madaan","year":"2022"},{"key":"2024050620131470100_bib85","article-title":"Self-refine: Iterative refinement with self-feedback","author":"Madaan","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib86","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-main.557","article-title":"Selfcheckgpt: Zero-resource black-box hallucination detection for generative large language models","author":"Manakul","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib87","article-title":"Flirt: Feedback loop in-context red teaming","author":"Mehrabi","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib88","doi-asserted-by":"publisher","first-page":"465","DOI":"10.1146\/annurev-psych-010416-044022","article-title":"Learning from errors","volume":"68","author":"Metcalfe","year":"2017","journal-title":"Annual Review of Psychology"},{"key":"2024050620131470100_bib89","article-title":"Selfcheck: Using LLMs to zero-shot check their own step-by-step reasoning","author":"Miao","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib90","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-main.741","article-title":"Factscore: Fine-grained atomic evaluation of factual precision in long form text generation","author":"Min","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib91","doi-asserted-by":"publisher","first-page":"11600","DOI":"10.18653\/v1\/2022.emnlp-main.797","article-title":"Fixing model bugs with natural language patches","volume-title":"Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing (EMNLP)","author":"Murty","year":"2022"},{"key":"2024050620131470100_bib92","doi-asserted-by":"publisher","first-page":"6591","DOI":"10.18653\/v1\/2023.emnlp-main.407","article-title":"MAF: Multi-aspect feedback for improving reasoning in large language models","volume-title":"Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing (EMNLP)","author":"Nathani","year":"2023"},{"key":"2024050620131470100_bib93","article-title":"LEVER: Learning to verify language-to-code generation with execution","volume-title":"Proceedings of the 40th International Conference on Machine Learning (ICML)","author":"Ni","year":"2023"},{"key":"2024050620131470100_bib94","article-title":"Demystifying GPT self-repair for code generation","author":"Olausson","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib95","doi-asserted-by":"publisher","first-page":"5469","DOI":"10.18653\/v1\/2023.acl-long.300","article-title":"Can lms learn new entities from descriptions? Challenges in propagating injected knowledge","volume-title":"Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (ACL)","author":"Onoe","year":"2023"},{"key":"2024050620131470100_bib96","unstructured":"OpenAI. 2023. GPT-4 technical report. CoRR, abs\/2303.08774."},{"key":"2024050620131470100_bib97","article-title":"Training language models to follow instructions with human feedback","volume-title":"Proceedings of the Annual Conference on Neural Information Processing Systems (NeurIPS)","author":"Ouyang","year":"2022"},{"key":"2024050620131470100_bib98","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.findings-emnlp.248","article-title":"Logic-LM: Empowering large language models with symbolic solvers for faithful logical reasoning","author":"Pan","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib99","article-title":"Language model self-improvement by reinforcement learning contemplation","author":"Pang","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib100","article-title":"REFINER: Reasoning feedback on intermediate representations","author":"Paul","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib101","article-title":"Check your facts and try again: Improving large language models with external knowledge and automated feedback","author":"Peng","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib102","first-page":"1","article-title":"Chatgpt vs human-authored text: Insights into controllable text summarization and sentence style transfer","volume-title":"Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics: Student Research Workshop (ACL)","author":"Dongqi","year":"2023"},{"key":"2024050620131470100_bib103","article-title":"Is chatgpt a general-purpose natural language processing task solver?","author":"Qin","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib104","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.findings-emnlp.804","article-title":"Leveraging GPT-4 for automatic translation post-editing","author":"Raunak","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib105","article-title":"STREET: A multi-task structured reasoning and explanation benchmark","volume-title":"Proceedings of the 11th International Conference on Learning Representations (ICLR)","author":"Ribeiro","year":"2023"},{"key":"2024050620131470100_bib106","doi-asserted-by":"publisher","first-page":"192","DOI":"10.18653\/v1\/S18-2024","article-title":"Quality signals in generated stories","volume-title":"Proceedings of the Seventh Joint Conference on Lexical and Computational Semantics (SEM@NAACL-HLT 2018)","author":"Sagarkar","year":"2018"},{"key":"2024050620131470100_bib107","doi-asserted-by":"publisher","first-page":"122","DOI":"10.18653\/v1\/2020.emnlp-main.9","article-title":"PRover: Proof generation for interpretable reasoning over rules","volume-title":"Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)","author":"Saha","year":"2020"},{"key":"2024050620131470100_bib108","article-title":"Self-critiquing models for assisting human evaluators","author":"Saunders","year":"2022","journal-title":"CoRR"},{"key":"2024050620131470100_bib109","doi-asserted-by":"publisher","first-page":"3053","DOI":"10.18653\/v1\/2023.eacl-main.223","article-title":"PECO: Examining single sentence label leakage in natural language inference datasets through progressive evaluation of cluster outliers","volume-title":"Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics (EACL)","author":"Saxon","year":"2023"},{"key":"2024050620131470100_bib110","article-title":"Training language models with language feedback at scale","author":"Scheurer","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib111","article-title":"PEER: A collaborative language model","volume-title":"Proceedings of the 11th International Conference on Learning Representations (ICLR)","author":"Schick","year":"2023"},{"key":"2024050620131470100_bib112","article-title":"Proximal policy optimization algorithms","author":"Schulman","year":"2017","journal-title":"CoRR"},{"key":"2024050620131470100_bib113","doi-asserted-by":"publisher","first-page":"4454","DOI":"10.18653\/v1\/2023.acl-long.244","article-title":"On second thought, let\u2019s not think step by step! Bias and toxicity in zero-shot reasoning","volume-title":"Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (ACL)","author":"Shaikh","year":"2023"},{"key":"2024050620131470100_bib114","article-title":"Reflexion: Language agents with verbal reinforcement learning","author":"Shinn","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib115","article-title":"Editable neural networks","volume-title":"Proceedings of the 8th International Conference on Learning Representations (ICLR)","author":"Sinitsin","year":"2020"},{"key":"2024050620131470100_bib116","doi-asserted-by":"publisher","first-page":"4753","DOI":"10.18653\/v1\/2022.naacl-main.350","article-title":"Partial-input baselines show that NLI models can ignore context, but they don\u2019t.","volume-title":"Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT)","author":"Srikanth","year":"2022"},{"key":"2024050620131470100_bib117","article-title":"GPT-4 doesn\u2019t know it\u2019s wrong: An analysis of iterative prompting for reasoning problems","author":"Stechly","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib118","doi-asserted-by":"publisher","first-page":"13003","DOI":"10.18653\/v1\/2023.findings-acl.824","article-title":"Challenging big-bench tasks and whether chain-of-thought can solve them","volume-title":"Findings of the Association for Computational Linguistics: ACL 2023","author":"Suzgun","year":"2023"},{"key":"2024050620131470100_bib119","doi-asserted-by":"publisher","first-page":"3621","DOI":"10.18653\/v1\/2021.findings-acl.317","article-title":"ProofWriter: Generating implications, proofs, and abductive statements over natural language","volume-title":"Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021","author":"Tafjord","year":"2021"},{"key":"2024050620131470100_bib120","doi-asserted-by":"publisher","first-page":"2078","DOI":"10.18653\/v1\/2022.emnlp-main.134","article-title":"Entailer: Answering questions with faithful and truthful chains of reasoning","volume-title":"Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing (EMNLP)","author":"Tafjord","year":"2022"},{"key":"2024050620131470100_bib121","article-title":"Repairing neural networks by leaving the right past behind","volume-title":"Proceedings of the 2022 Annual Conference on Neural Information Processing Systems (NeurIPS)","author":"Tanno","year":"2022"},{"key":"2024050620131470100_bib122","article-title":"LLMs cannot find reasoning errors, but can correct them!","author":"Tyen","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib123","article-title":"Solving math word problems with process- and outcome-based feedback","author":"Uesato","year":"2022","journal-title":"CoRR"},{"key":"2024050620131470100_bib124","doi-asserted-by":"publisher","first-page":"915","DOI":"10.18653\/v1\/2021.acl-short.115","article-title":"Berttune: Fine-tuning neural machine translation with bertscore","volume-title":"Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (ACL\/IJCNLP)","author":"Unanue","year":"2021"},{"key":"2024050620131470100_bib125","article-title":"Can large language models really improve by self-critiquing their own plans?","volume":"abs\/2310.08118","author":"Valmeekam","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib126","article-title":"A stitch in time saves nine: Detecting and mitigating hallucinations of LLMs by validating low-confidence generation","volume":"abs\/2307.03987","author":"Varshney","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib127","doi-asserted-by":"publisher","first-page":"1010","DOI":"10.18653\/v1\/2022.naacl-main.74","article-title":"Factpegasus: Factuality-aware pre-training and fine-tuning for abstractive summarization","volume-title":"Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT)","author":"Wan","year":"2022"},{"key":"2024050620131470100_bib128","article-title":"Decodingtrust: A comprehensive assessment of trustworthiness in GPT models","author":"Wang","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib129","article-title":"Apollo\u2019s oracle: Retrieval-augmented reasoning in multi-agent debates","author":"Wang","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib130","article-title":"A comprehensive survey of continual learning: Theory, method and application","author":"Wang","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib131","doi-asserted-by":"publisher","first-page":"3859","DOI":"10.24963\/ijcai.2017\/539","article-title":"Predicting the quality of short narratives from social media","volume-title":"Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI)","author":"Wang","year":"2017"},{"key":"2024050620131470100_bib132","article-title":"Emergent abilities of large language models","author":"Wei","year":"2022","journal-title":"CoRR"},{"key":"2024050620131470100_bib133","article-title":"Chain-of-thought prompting elicits reasoning in large language models","volume-title":"Proceedings of the Annual Conference on Neural Information Processing Systems (NeurIPS)","author":"Wei","year":"2022"},{"key":"2024050620131470100_bib134","article-title":"Generating sequences by learning to self-correct","volume-title":"Proceedings of The 11th International Conference on Learning Representations (ICLR)","author":"Welleck","year":"2023"},{"key":"2024050620131470100_bib135","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.findings-emnlp.167","article-title":"Large language models are better reasoners with self-verification","author":"Weng","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib136","article-title":"Fine-grained human feedback gives better rewards for language model training","author":"Zeqiu","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib137","article-title":"Reasoning or reciting? Exploring the capabilities and limitations of language models through counterfactual tasks","author":"Zhaofeng","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib138","article-title":"Decomposition enhances reasoning via self-evaluation guided decoding","author":"Xie","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib139","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-main.365","article-title":"INSTRUCTSCORE: Towards explainable text generation evaluation with automatic feedback","author":"Wenda","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib140","article-title":"Sqlnet: Generating structured queries from natural language without reinforcement learning","author":"Xiaojun","year":"2017","journal-title":"CoRR"},{"key":"2024050620131470100_bib141","doi-asserted-by":"publisher","first-page":"3149","DOI":"10.18653\/v1\/2023.acl-long.177","article-title":"Learning to simulate natural language feedback for interactive semantic parsing","volume-title":"Proceedings of the 61th Annual Meeting of the Association for Computational Linguistics (ACL)","author":"Yan","year":"2023"},{"key":"2024050620131470100_bib142","doi-asserted-by":"publisher","first-page":"89","DOI":"10.18653\/v1\/2022.emnlp-main.7","article-title":"Generating natural language proofs with verifier-guided search","volume-title":"Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing (EMNLP)","author":"Yang","year":"2022"},{"key":"2024050620131470100_bib143","doi-asserted-by":"publisher","first-page":"3511","DOI":"10.18653\/v1\/2021.naacl-main.276","article-title":"FUDGE: Controlled text generation with future discriminators","volume-title":"Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT)","author":"Yang","year":"2021"},{"key":"2024050620131470100_bib144","doi-asserted-by":"publisher","first-page":"4393","DOI":"10.18653\/v1\/2022.emnlp-main.296","article-title":"Re3: Generating longer stories with recursive reprompting and revision","volume-title":"Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing (EMNLP)","author":"Yang","year":"2022"},{"key":"2024050620131470100_bib145","article-title":"Tree of thoughts: Deliberate problem solving with large language models","author":"Yao","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib146","article-title":"Editing large language models: Problems, methods, and opportunities","author":"Yao","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib147","article-title":"Selfee: Iterative self-revising LLM empowered by self-feedback generation","author":"Ye","year":"2023"},{"key":"2024050620131470100_bib148","doi-asserted-by":"publisher","first-page":"1653","DOI":"10.18653\/v1\/D18-1193","article-title":"SyntaxSQLNet: Syntax tree networks for complex and cross-domain text-to-SQL task","volume-title":"Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing (EMNLP)","author":"Tao","year":"2018"},{"key":"2024050620131470100_bib149","article-title":"Improving language models via plug-and-play retrieval feedback","author":"Wenhao","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib150","article-title":"System-level natural language feedback","author":"Yuan","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib151","article-title":"Star: Bootstrapping reasoning with reasoning","volume-title":"Proceedings of the Annual Conference on Neural Information Processing Systems (NeurIPS)","author":"Zelikman","year":"2022"},{"key":"2024050620131470100_bib152","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.acl-long.45","article-title":"Self-edit: Fault-aware code editor for code generation","author":"Zhang","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib153","article-title":"Algo: Synthesizing algorithmic programs with generated oracle verifiers","author":"Zhang","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib154","article-title":"How language model hallucinations can snowball","author":"Zhang","year":"2023","journal-title":"CoRR"},{"key":"2024050620131470100_bib155","doi-asserted-by":"publisher","first-page":"4471","DOI":"10.18653\/v1\/2023.acl-long.245","article-title":"Solving math word problems via cooperative reasoning induced language models","volume-title":"Processings of the 61th Annual Meeting of the Association for Computational Linguistics (ACL)","author":"Zhu","year":"2023"},{"key":"2024050620131470100_bib156","article-title":"Red teaming chatgpt via jailbreaking: Bias, robustness, reliability and toxicity","author":"Zhuo","year":"2023","journal-title":"CoRR"}],"container-title":["Transactions of the Association for Computational Linguistics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/direct.mit.edu\/tacl\/article-pdf\/doi\/10.1162\/tacl_a_00660\/2369509\/tacl_a_00660.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/direct.mit.edu\/tacl\/article-pdf\/doi\/10.1162\/tacl_a_00660\/2369509\/tacl_a_00660.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,5,6]],"date-time":"2024-05-06T16:14:17Z","timestamp":1715012057000},"score":1,"resource":{"primary":{"URL":"https:\/\/direct.mit.edu\/tacl\/article\/doi\/10.1162\/tacl_a_00660\/120911\/Automatically-Correcting-Large-Language-Models"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024]]},"references-count":156,"URL":"https:\/\/doi.org\/10.1162\/tacl_a_00660","relation":{},"ISSN":["2307-387X"],"issn-type":[{"value":"2307-387X","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2024]]},"published":{"date-parts":[[2024]]}}}