{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:ZZVAMWKQ7JYI5A2S2LC3LNFVND","short_pith_number":"pith:ZZVAMWKQ","schema_version":"1.0","canonical_sha256":"ce6a065950fa708e8352d2c5b5b4b568dd60b9f997be3840304ea6cc3727f204","source":{"kind":"arxiv","id":"2310.20689","version":4},"attestation_state":"computed","paper":{"title":"Learning From Mistakes Makes LLM Better Reasoner","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Jian-Guang Lou, Nanning Zheng, Shengnan An, Weizhu Chen, Zeqi Lin, Zexiong Ma","submitted_at":"2023-10-31T17:52:22Z","abstract_excerpt":"Large language models (LLMs) recently exhibited remarkable reasoning capabilities on solving math problems. To further improve their reasoning capabilities, this work explores whether LLMs can LEarn from MistAkes (LEMA), akin to the human learning process. Consider a human student who failed to solve a math problem, he will learn from what mistake he has made and how to correct it. Mimicking this error-driven learning process, LEMA incorporates mistake-correction data pairs during fine-tuning LLMs. Specifically, we first collect inaccurate reasoning paths from various LLMs, and then employ GPT"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2310.20689","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-10-31T17:52:22Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"81464295c7a392c69e5a559ac04c3f6fe4204380fd0a2879eef66fceb127c374","abstract_canon_sha256":"6c52c0c5e679c6a57c1cea142208d4bb2e291937a17a2ca51b5bbedae6bbab5c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:02:02.657439Z","signature_b64":"6N+7uLJa9hrjxyI9iyWl9cCsaqGUTOv9OKBMdm5zRpkYMXEgyAqNqyIwtfi6rcyAmf7oTXx8KrMu8koJ0yf8Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ce6a065950fa708e8352d2c5b5b4b568dd60b9f997be3840304ea6cc3727f204","last_reissued_at":"2026-07-05T08:02:02.656898Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:02:02.656898Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning From Mistakes Makes LLM Better Reasoner","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Jian-Guang Lou, Nanning Zheng, Shengnan An, Weizhu Chen, Zeqi Lin, Zexiong Ma","submitted_at":"2023-10-31T17:52:22Z","abstract_excerpt":"Large language models (LLMs) recently exhibited remarkable reasoning capabilities on solving math problems. To further improve their reasoning capabilities, this work explores whether LLMs can LEarn from MistAkes (LEMA), akin to the human learning process. Consider a human student who failed to solve a math problem, he will learn from what mistake he has made and how to correct it. Mimicking this error-driven learning process, LEMA incorporates mistake-correction data pairs during fine-tuning LLMs. Specifically, we first collect inaccurate reasoning paths from various LLMs, and then employ GPT"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.20689","kind":"arxiv","version":4},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2310.20689/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2310.20689","created_at":"2026-07-05T08:02:02.656959+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.20689v4","created_at":"2026-07-05T08:02:02.656959+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.20689","created_at":"2026-07-05T08:02:02.656959+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZZVAMWKQ7JYI","created_at":"2026-07-05T08:02:02.656959+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZZVAMWKQ7JYI5A2S","created_at":"2026-07-05T08:02:02.656959+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZZVAMWKQ","created_at":"2026-07-05T08:02:02.656959+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.06065","citing_title":"SWE-Review: Closing the Loop on Issue Resolution with Agentic Code Review","ref_index":3,"is_internal_anchor":true},{"citing_arxiv_id":"2410.13181","citing_title":"AdaSwitch: Adaptive Switching between Small and Large Agents for Effective Cloud-Local Collaborative Learning","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2509.03540","citing_title":"Improving Factuality in LLMs via Inference-Time Knowledge Graph Construction","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2402.13228","citing_title":"Smaug: Fixing Failure Modes of Preference Optimisation with DPO-Positive","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14539","citing_title":"Learning from Failures: Correction-Oriented Policy Optimization with Verifiable Rewards","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2402.02716","citing_title":"Understanding the planning of LLM agents: A survey","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2502.17419","citing_title":"From System 1 to System 2: A Survey of Reasoning Large Language Models","ref_index":226,"is_internal_anchor":false},{"citing_arxiv_id":"2503.09567","citing_title":"Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Models","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZZVAMWKQ7JYI5A2S2LC3LNFVND","json":"https://pith.science/pith/ZZVAMWKQ7JYI5A2S2LC3LNFVND.json","graph_json":"https://pith.science/api/pith-number/ZZVAMWKQ7JYI5A2S2LC3LNFVND/graph.json","events_json":"https://pith.science/api/pith-number/ZZVAMWKQ7JYI5A2S2LC3LNFVND/events.json","paper":"https://pith.science/paper/ZZVAMWKQ"},"agent_actions":{"view_html":"https://pith.science/pith/ZZVAMWKQ7JYI5A2S2LC3LNFVND","download_json":"https://pith.science/pith/ZZVAMWKQ7JYI5A2S2LC3LNFVND.json","view_paper":"https://pith.science/paper/ZZVAMWKQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.20689&json=true","fetch_graph":"https://pith.science/api/pith-number/ZZVAMWKQ7JYI5A2S2LC3LNFVND/graph.json","fetch_events":"https://pith.science/api/pith-number/ZZVAMWKQ7JYI5A2S2LC3LNFVND/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZZVAMWKQ7JYI5A2S2LC3LNFVND/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZZVAMWKQ7JYI5A2S2LC3LNFVND/action/storage_attestation","attest_author":"https://pith.science/pith/ZZVAMWKQ7JYI5A2S2LC3LNFVND/action/author_attestation","sign_citation":"https://pith.science/pith/ZZVAMWKQ7JYI5A2S2LC3LNFVND/action/citation_signature","submit_replication":"https://pith.science/pith/ZZVAMWKQ7JYI5A2S2LC3LNFVND/action/replication_record"}},"created_at":"2026-07-05T08:02:02.656959+00:00","updated_at":"2026-07-05T08:02:02.656959+00:00"}