{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:XP5QRJWORXVGYL5LSNZ3LPVQYR","short_pith_number":"pith:XP5QRJWO","schema_version":"1.0","canonical_sha256":"bbfb08a6ce8dea6c2fab9373b5beb0c471ac5638b82199949a8de7ac157dbfa3","source":{"kind":"arxiv","id":"2510.18814","version":3},"attestation_state":"computed","paper":{"title":"A Model Can Help Itself: Reward-Free Self-Training for LLM Reasoning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Language models can improve their reasoning by training on responses they generate themselves.","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Anthony Man-Cho So, Lei Zhao, Mengqi Li, Ruoyu Sun, Xiao Li","submitted_at":"2025-10-21T17:15:56Z","abstract_excerpt":"Can language models improve their reasoning performance without external rewards, using only their own sampled responses for training? We show that they can. We propose Self-evolving Post-Training (SePT), a simple post-training method that alternates between self-generation and training on self-generated responses. It repeatedly samples questions, uses the model itself to generate responses under a specified sampling temperature, and then trains the model on the self-generated data. In this self-training loop, we use an online data refresh mechanism, where each new batch is generated by the mo"},"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":true},"canonical_record":{"source":{"id":"2510.18814","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-10-21T17:15:56Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3da41121e66a06f632b807a7bacb8a67fc88c090f8e70453dc0c1151ff1a4d99","abstract_canon_sha256":"a2a6d74918e4bf28b16bc80f1d5e27a016ac55827482ebac9d171647432ca76e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:00:26.366372Z","signature_b64":"cwC3wfbzGfGfP6qgYXAuD2g7qmyeNg1/Qwk/Q7+CEi9L4hYBVxwID3tCQEq0xS12r8pcABoGihILaQsfa1mBAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bbfb08a6ce8dea6c2fab9373b5beb0c471ac5638b82199949a8de7ac157dbfa3","last_reissued_at":"2026-05-20T00:00:26.365399Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:00:26.365399Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Model Can Help Itself: Reward-Free Self-Training for LLM Reasoning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Language models can improve their reasoning by training on responses they generate themselves.","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Anthony Man-Cho So, Lei Zhao, Mengqi Li, Ruoyu Sun, Xiao Li","submitted_at":"2025-10-21T17:15:56Z","abstract_excerpt":"Can language models improve their reasoning performance without external rewards, using only their own sampled responses for training? We show that they can. We propose Self-evolving Post-Training (SePT), a simple post-training method that alternates between self-generation and training on self-generated responses. It repeatedly samples questions, uses the model itself to generate responses under a specified sampling temperature, and then trains the model on the self-generated data. In this self-training loop, we use an online data refresh mechanism, where each new batch is generated by the mo"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We show that they can. We propose Self-evolving Post-Training (SePT), a simple post-training method that alternates between self-generation and training on self-generated responses. Across six math reasoning benchmarks, SePT improves a strong no-training baseline... and in some settings can even approach the performance of Reinforcement Learning with Verifiable Rewards (RLVR).","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That self-generated responses supply a net positive training signal rather than reinforcing the model's existing errors or hallucinations, which must hold for the iterative self-training loop to produce sustained gains without external verification or filtering.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"SePT enables LLMs to improve math reasoning on multiple benchmarks by iteratively training on their own low-temperature generated responses using an online data refresh mechanism.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Language models can improve their reasoning by training on responses they generate themselves.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"d1eefac338f9d4b3286071d0d1691c85c584bb514eff26205aa22a72cc14b1da"},"source":{"id":"2510.18814","kind":"arxiv","version":3},"verdict":{"id":"f4d25a43-0b64-42de-9fb7-be4d16c84530","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-18T05:07:29.377760Z","strongest_claim":"We show that they can. We propose Self-evolving Post-Training (SePT), a simple post-training method that alternates between self-generation and training on self-generated responses. Across six math reasoning benchmarks, SePT improves a strong no-training baseline... and in some settings can even approach the performance of Reinforcement Learning with Verifiable Rewards (RLVR).","one_line_summary":"SePT enables LLMs to improve math reasoning on multiple benchmarks by iteratively training on their own low-temperature generated responses using an online data refresh mechanism.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That self-generated responses supply a net positive training signal rather than reinforcing the model's existing errors or hallucinations, which must hold for the iterative self-training loop to produce sustained gains without external verification or filtering.","pith_extraction_headline":"Language models can improve their reasoning by training on responses they generate themselves."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2510.18814/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":2,"snapshot_sha256":"1da77da5bf2e75e62ba903468a8c56f50941e63395d2864ab68fa7a509218b65"},"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":"2510.18814","created_at":"2026-05-20T00:00:26.365554+00:00"},{"alias_kind":"arxiv_version","alias_value":"2510.18814v3","created_at":"2026-05-20T00:00:26.365554+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2510.18814","created_at":"2026-05-20T00:00:26.365554+00:00"},{"alias_kind":"pith_short_12","alias_value":"XP5QRJWORXVG","created_at":"2026-05-20T00:00:26.365554+00:00"},{"alias_kind":"pith_short_16","alias_value":"XP5QRJWORXVGYL5L","created_at":"2026-05-20T00:00:26.365554+00:00"},{"alias_kind":"pith_short_8","alias_value":"XP5QRJWO","created_at":"2026-05-20T00:00:26.365554+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2605.06650","citing_title":"Beyond Negative Rollouts: Positive-Only Policy Optimization with Implicit Negative Gradients","ref_index":55,"is_internal_anchor":true},{"citing_arxiv_id":"2605.04542","citing_title":"Power Distribution Bridges Sampling, Self-Reward RL, and Self-Distillation","ref_index":198,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XP5QRJWORXVGYL5LSNZ3LPVQYR","json":"https://pith.science/pith/XP5QRJWORXVGYL5LSNZ3LPVQYR.json","graph_json":"https://pith.science/api/pith-number/XP5QRJWORXVGYL5LSNZ3LPVQYR/graph.json","events_json":"https://pith.science/api/pith-number/XP5QRJWORXVGYL5LSNZ3LPVQYR/events.json","paper":"https://pith.science/paper/XP5QRJWO"},"agent_actions":{"view_html":"https://pith.science/pith/XP5QRJWORXVGYL5LSNZ3LPVQYR","download_json":"https://pith.science/pith/XP5QRJWORXVGYL5LSNZ3LPVQYR.json","view_paper":"https://pith.science/paper/XP5QRJWO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2510.18814&json=true","fetch_graph":"https://pith.science/api/pith-number/XP5QRJWORXVGYL5LSNZ3LPVQYR/graph.json","fetch_events":"https://pith.science/api/pith-number/XP5QRJWORXVGYL5LSNZ3LPVQYR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XP5QRJWORXVGYL5LSNZ3LPVQYR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XP5QRJWORXVGYL5LSNZ3LPVQYR/action/storage_attestation","attest_author":"https://pith.science/pith/XP5QRJWORXVGYL5LSNZ3LPVQYR/action/author_attestation","sign_citation":"https://pith.science/pith/XP5QRJWORXVGYL5LSNZ3LPVQYR/action/citation_signature","submit_replication":"https://pith.science/pith/XP5QRJWORXVGYL5LSNZ3LPVQYR/action/replication_record"}},"created_at":"2026-05-20T00:00:26.365554+00:00","updated_at":"2026-05-20T00:00:26.365554+00:00"}