{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:AE72JTNISJ6LLFHXJHFFTFDCIE","short_pith_number":"pith:AE72JTNI","schema_version":"1.0","canonical_sha256":"013fa4cda8927cb594f749ca599462411f4d30e4c148d2ebf16bfe3d173a40e0","source":{"kind":"arxiv","id":"2605.18141","version":1},"attestation_state":"computed","paper":{"title":"A Brief Overview: On-Policy Self-Distillation In Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.HC","authors_text":"Fangming Cui, Jiahong Li, Sunan Li","submitted_at":"2026-05-18T09:47:53Z","abstract_excerpt":"On-Policy Self-Distillation (OPSD) introduces a unified learning framework in which a single large language model simultaneously serves as both teacher and student. Unlike conventional knowledge distillation that relies on a separate, often larger teacher model, OPSD operates under different contextual roles: the teacher policy is granted privileged access to verified reasoning traces, while the student policy observes only the problem statement. OPSD is trained to minimize per-token distributional divergence between the two roles over trajectories sampled from the student itself, thereby alig"},"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":"2605.18141","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.HC","submitted_at":"2026-05-18T09:47:53Z","cross_cats_sorted":[],"title_canon_sha256":"5bd2a7d1d6cbc177c7d77db1aad874b87c04766a25ebdd26ad5b3cc565e3c634","abstract_canon_sha256":"6dbe91da23fcc0ff68aca40f9505613b067ede0b18063eb6d0d2b57502167cd8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:05:18.429352Z","signature_b64":"JFYM6351DYUufjkIBkYBuHXCRJ11E2H0FFcIdFUzxqlbAxoDbuZizlRdIPpJa2KrB4as8KM2FTAJGTKcQmldDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"013fa4cda8927cb594f749ca599462411f4d30e4c148d2ebf16bfe3d173a40e0","last_reissued_at":"2026-05-20T00:05:18.428447Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:05:18.428447Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Brief Overview: On-Policy Self-Distillation In Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.HC","authors_text":"Fangming Cui, Jiahong Li, Sunan Li","submitted_at":"2026-05-18T09:47:53Z","abstract_excerpt":"On-Policy Self-Distillation (OPSD) introduces a unified learning framework in which a single large language model simultaneously serves as both teacher and student. Unlike conventional knowledge distillation that relies on a separate, often larger teacher model, OPSD operates under different contextual roles: the teacher policy is granted privileged access to verified reasoning traces, while the student policy observes only the problem statement. OPSD is trained to minimize per-token distributional divergence between the two roles over trajectories sampled from the student itself, thereby alig"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.18141","kind":"arxiv","version":1},"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/2605.18141/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-19T23:41:59.108959Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T23:33:35.382558Z","status":"skipped","version":"1.0.0","findings_count":0}],"snapshot_sha256":"5bdc4bf723982a07f976422a43f5d99244a2fce0d459e860f0b8fa77e7305339"},"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":"2605.18141","created_at":"2026-05-20T00:05:18.428583+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.18141v1","created_at":"2026-05-20T00:05:18.428583+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.18141","created_at":"2026-05-20T00:05:18.428583+00:00"},{"alias_kind":"pith_short_12","alias_value":"AE72JTNISJ6L","created_at":"2026-05-20T00:05:18.428583+00:00"},{"alias_kind":"pith_short_16","alias_value":"AE72JTNISJ6LLFHX","created_at":"2026-05-20T00:05:18.428583+00:00"},{"alias_kind":"pith_short_8","alias_value":"AE72JTNI","created_at":"2026-05-20T00:05:18.428583+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AE72JTNISJ6LLFHXJHFFTFDCIE","json":"https://pith.science/pith/AE72JTNISJ6LLFHXJHFFTFDCIE.json","graph_json":"https://pith.science/api/pith-number/AE72JTNISJ6LLFHXJHFFTFDCIE/graph.json","events_json":"https://pith.science/api/pith-number/AE72JTNISJ6LLFHXJHFFTFDCIE/events.json","paper":"https://pith.science/paper/AE72JTNI"},"agent_actions":{"view_html":"https://pith.science/pith/AE72JTNISJ6LLFHXJHFFTFDCIE","download_json":"https://pith.science/pith/AE72JTNISJ6LLFHXJHFFTFDCIE.json","view_paper":"https://pith.science/paper/AE72JTNI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.18141&json=true","fetch_graph":"https://pith.science/api/pith-number/AE72JTNISJ6LLFHXJHFFTFDCIE/graph.json","fetch_events":"https://pith.science/api/pith-number/AE72JTNISJ6LLFHXJHFFTFDCIE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AE72JTNISJ6LLFHXJHFFTFDCIE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AE72JTNISJ6LLFHXJHFFTFDCIE/action/storage_attestation","attest_author":"https://pith.science/pith/AE72JTNISJ6LLFHXJHFFTFDCIE/action/author_attestation","sign_citation":"https://pith.science/pith/AE72JTNISJ6LLFHXJHFFTFDCIE/action/citation_signature","submit_replication":"https://pith.science/pith/AE72JTNISJ6LLFHXJHFFTFDCIE/action/replication_record"}},"created_at":"2026-05-20T00:05:18.428583+00:00","updated_at":"2026-05-20T00:05:18.428583+00:00"}