{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:FJ7SS4UWAI7H4H3KT5VTQMNL4J","short_pith_number":"pith:FJ7SS4UW","schema_version":"1.0","canonical_sha256":"2a7f297296023e7e1f6a9f6b3831abe26735212df6938c89631204fb7806d50d","source":{"kind":"arxiv","id":"2502.03387","version":3},"attestation_state":"computed","paper":{"title":"LIMO: Less is More for Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Sophisticated mathematical reasoning emerges in large language models from only a few strategically designed examples.","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ethan Chern, Pengfei Liu, Shijie Xia, Yang Xiao, Yixin Ye, Zhen Huang","submitted_at":"2025-02-05T17:23:45Z","abstract_excerpt":"We challenge the prevailing assumption that complex reasoning in large language models (LLMs) necessitates massive training data. We demonstrate that sophisticated mathematical reasoning can emerge with only a few examples. Specifically, through simple supervised fine-tuning, our model, LIMO, achieves 63.3\\% accuracy on AIME24 and 95.6\\% on MATH500, surpassing previous fine-tuned models (6.5\\% on AIME24, 59.2\\% on MATH500) while using only 1\\% of the training data required by prior approaches. Furthermore, LIMO exhibits strong out-of-distribution generalization, achieving a 45.8\\% absolute imp"},"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":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2502.03387","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-05T17:23:45Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b2893585327c5dea96c02e8be12b73ff54b659b0d8de6de73f0599ec12965715","abstract_canon_sha256":"348bddcd4634c3d1407c8e87fe25abf1925e655f5771c0cad444c8e714a16eaf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:38:15.429538Z","signature_b64":"MZztHjWr3KxccnlUn2Xy/vkmNQFv5F1/edfQ7lY7Xp9XEYFEzOhehBVdDU1wl+1/wvPwM8MoNEHEUeK5Vn8MDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2a7f297296023e7e1f6a9f6b3831abe26735212df6938c89631204fb7806d50d","last_reissued_at":"2026-05-17T23:38:15.428774Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:38:15.428774Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LIMO: Less is More for Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Sophisticated mathematical reasoning emerges in large language models from only a few strategically designed examples.","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ethan Chern, Pengfei Liu, Shijie Xia, Yang Xiao, Yixin Ye, Zhen Huang","submitted_at":"2025-02-05T17:23:45Z","abstract_excerpt":"We challenge the prevailing assumption that complex reasoning in large language models (LLMs) necessitates massive training data. We demonstrate that sophisticated mathematical reasoning can emerge with only a few examples. Specifically, through simple supervised fine-tuning, our model, LIMO, achieves 63.3\\% accuracy on AIME24 and 95.6\\% on MATH500, surpassing previous fine-tuned models (6.5\\% on AIME24, 59.2\\% on MATH500) while using only 1\\% of the training data required by prior approaches. Furthermore, LIMO exhibits strong out-of-distribution generalization, achieving a 45.8\\% absolute imp"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"sophisticated mathematical reasoning can emerge with only a few examples... through minimal but strategically designed demonstrations of cognitive processes.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The foundation model has already comprehensively encoded the relevant domain knowledge during pre-training, and the small set of post-training examples functions as effective cognitive templates.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"LIMO achieves 63.3% on AIME24 and 95.6% on MATH500 via supervised fine-tuning on roughly 1% of the data used by prior models, supporting the claim that minimal strategic examples suffice when pre-training has already encoded domain knowledge.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Sophisticated mathematical reasoning emerges in large language models from only a few strategically designed examples.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"16afdebc6eff494daf1db31257ec9dfe30625c4e5b5eb445ffcaf4cdab9472e7"},"source":{"id":"2502.03387","kind":"arxiv","version":3},"verdict":{"id":"5e2233a9-6c9d-4a39-967b-ea9e5e9f5b15","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-17T02:06:10.598649Z","strongest_claim":"sophisticated mathematical reasoning can emerge with only a few examples... through minimal but strategically designed demonstrations of cognitive processes.","one_line_summary":"LIMO achieves 63.3% on AIME24 and 95.6% on MATH500 via supervised fine-tuning on roughly 1% of the data used by prior models, supporting the claim that minimal strategic examples suffice when pre-training has already encoded domain knowledge.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The foundation model has already comprehensively encoded the relevant domain knowledge during pre-training, and the small set of post-training examples functions as effective cognitive templates.","pith_extraction_headline":"Sophisticated mathematical reasoning emerges in large language models from only a few strategically designed examples."},"references":{"count":299,"sample":[{"doi":"","year":null,"title":"Attention is All you Need , url =","work_id":"fa4d7178-75f8-4139-9b39-41030a2917db","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"OpenAI o1 System Card , author=. 2024 , eprint=","work_id":"f4926faa-1e08-4b8f-8e4e-7065c75fd9fe","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training , author=. 2025 , eprint=","work_id":"e4119e9f-6591-40a0-b465-9eb7781d4d57","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"Benchmarking Benchmark Leakage in Large Language Models , author=. 2024 , eprint=","work_id":"de8590fb-4f72-4557-b70e-22316fbe659b","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"MathPile: A Billion-Token-Scale Pretraining Corpus for Math , author=. 2024 , eprint=","work_id":"9d11acdc-ea9d-4f90-9f36-42a7ab665dac","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":299,"snapshot_sha256":"70fcbfc1f59e5910283da5046d7d3424edeb0767511024fd422a90bdf915597f","internal_anchors":38},"formal_canon":{"evidence_count":2,"snapshot_sha256":"1f15360cc5861bd2ab3938d5626396d8d049005b3998cd46bd6f7f566eab11e3"},"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":"2502.03387","created_at":"2026-05-17T23:38:15.428890+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.03387v3","created_at":"2026-05-17T23:38:15.428890+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.03387","created_at":"2026-05-17T23:38:15.428890+00:00"},{"alias_kind":"pith_short_12","alias_value":"FJ7SS4UWAI7H","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"FJ7SS4UWAI7H4H3K","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"FJ7SS4UW","created_at":"2026-05-18T12:33:37.589309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":34,"internal_anchor_count":34,"sample":[{"citing_arxiv_id":"2605.23398","citing_title":"TPMM-DPO: Trajectory-aware Preference-guided Model Merging for Iterative Direct Preference Optimization","ref_index":30,"is_internal_anchor":true},{"citing_arxiv_id":"2504.01990","citing_title":"Advances and Challenges in Foundation Agents: From Brain-Inspired Intelligence to Evolutionary, Collaborative, and Safe Systems","ref_index":87,"is_internal_anchor":true},{"citing_arxiv_id":"2504.02181","citing_title":"A Survey of Scaling in Large Language Model Reasoning","ref_index":243,"is_internal_anchor":true},{"citing_arxiv_id":"2605.11739","citing_title":"Learning to Foresee: Unveiling the Unlocking Efficiency of On-Policy Distillation","ref_index":25,"is_internal_anchor":true},{"citing_arxiv_id":"2510.18814","citing_title":"A Model Can Help Itself: Reward-Free Self-Training for LLM Reasoning","ref_index":23,"is_internal_anchor":true},{"citing_arxiv_id":"2602.01970","citing_title":"Small Generalizable Prompt Predictive Models Can Steer Efficient RL Post-Training of Large Reasoning Models","ref_index":40,"is_internal_anchor":true},{"citing_arxiv_id":"2602.07832","citing_title":"rePIRL: Learn PRM with Inverse RL for LLM Reasoning","ref_index":35,"is_internal_anchor":true},{"citing_arxiv_id":"2506.13351","citing_title":"Direct Reasoning Optimization: Token-Level Reasoning Reflectivity Meets Rubric Gates for Unverifiable Tasks","ref_index":30,"is_internal_anchor":true},{"citing_arxiv_id":"2507.14958","citing_title":"MUR: Momentum Uncertainty guided Reasoning for Large Language Models","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"2507.21166","citing_title":"The Ratchet Effect in Silico through Interaction-Driven Cumulative Intelligence in Large Language Models","ref_index":54,"is_internal_anchor":true},{"citing_arxiv_id":"2508.15815","citing_title":"User-Assistant Bias in LLMs","ref_index":21,"is_internal_anchor":true},{"citing_arxiv_id":"2508.15202","citing_title":"Fin-PRM: A Domain-Specialized Process Reward Model for Financial Reasoning in Large Language Models","ref_index":26,"is_internal_anchor":true},{"citing_arxiv_id":"2508.17784","citing_title":"Proximal Supervised Fine-Tuning","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2510.04265","citing_title":"Don't Pass@k: A Bayesian Framework for Large Language Model Evaluation","ref_index":89,"is_internal_anchor":true},{"citing_arxiv_id":"2510.18814","citing_title":"A Model Can Help Itself: Reward-Free Self-Training for LLM Reasoning","ref_index":23,"is_internal_anchor":true},{"citing_arxiv_id":"2510.19669","citing_title":"DiffAdapt: Difficulty-Adaptive Reasoning for Token-Efficient LLM Inference","ref_index":18,"is_internal_anchor":true},{"citing_arxiv_id":"2511.07940","citing_title":"ISExplore:Informative Segment Selection for Efficient Personalized 3D Talking Face Generation","ref_index":35,"is_internal_anchor":true},{"citing_arxiv_id":"2507.02592","citing_title":"WebSailor: Navigating Super-human Reasoning for Web Agent","ref_index":30,"is_internal_anchor":true},{"citing_arxiv_id":"2512.11470","citing_title":"Rethinking Expert Trajectory Utilization in LLM Post-training for Mathematical Reasoning","ref_index":46,"is_internal_anchor":true},{"citing_arxiv_id":"2601.13262","citing_title":"CURE-Med: Curriculum-Informed Reinforcement Learning for Multilingual Medical Reasoning","ref_index":30,"is_internal_anchor":true},{"citing_arxiv_id":"2601.14249","citing_title":"Which Reasoning Trajectories Teach Students to Reason Better? A Simple Metric of Informative Alignment","ref_index":45,"is_internal_anchor":true},{"citing_arxiv_id":"2601.21619","citing_title":"On the Overscaling Curse of Parallel Thinking: System Efficacy Contradicts Sample Efficiency","ref_index":33,"is_internal_anchor":true},{"citing_arxiv_id":"2504.14945","citing_title":"Learning to Reason under Off-Policy Guidance","ref_index":37,"is_internal_anchor":true},{"citing_arxiv_id":"2503.12605","citing_title":"Multimodal Chain-of-Thought Reasoning: A Comprehensive Survey","ref_index":281,"is_internal_anchor":true},{"citing_arxiv_id":"2605.12906","citing_title":"Data Difficulty and the Generalization--Extrapolation Tradeoff in LLM Fine-Tuning","ref_index":17,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FJ7SS4UWAI7H4H3KT5VTQMNL4J","json":"https://pith.science/pith/FJ7SS4UWAI7H4H3KT5VTQMNL4J.json","graph_json":"https://pith.science/api/pith-number/FJ7SS4UWAI7H4H3KT5VTQMNL4J/graph.json","events_json":"https://pith.science/api/pith-number/FJ7SS4UWAI7H4H3KT5VTQMNL4J/events.json","paper":"https://pith.science/paper/FJ7SS4UW"},"agent_actions":{"view_html":"https://pith.science/pith/FJ7SS4UWAI7H4H3KT5VTQMNL4J","download_json":"https://pith.science/pith/FJ7SS4UWAI7H4H3KT5VTQMNL4J.json","view_paper":"https://pith.science/paper/FJ7SS4UW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.03387&json=true","fetch_graph":"https://pith.science/api/pith-number/FJ7SS4UWAI7H4H3KT5VTQMNL4J/graph.json","fetch_events":"https://pith.science/api/pith-number/FJ7SS4UWAI7H4H3KT5VTQMNL4J/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FJ7SS4UWAI7H4H3KT5VTQMNL4J/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FJ7SS4UWAI7H4H3KT5VTQMNL4J/action/storage_attestation","attest_author":"https://pith.science/pith/FJ7SS4UWAI7H4H3KT5VTQMNL4J/action/author_attestation","sign_citation":"https://pith.science/pith/FJ7SS4UWAI7H4H3KT5VTQMNL4J/action/citation_signature","submit_replication":"https://pith.science/pith/FJ7SS4UWAI7H4H3KT5VTQMNL4J/action/replication_record"}},"created_at":"2026-05-17T23:38:15.428890+00:00","updated_at":"2026-05-17T23:38:15.428890+00:00"}