{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:7XEA25FQG2QRI2FNXXF7MDASVD","short_pith_number":"pith:7XEA25FQ","schema_version":"1.0","canonical_sha256":"fdc80d74b036a11468adbdcbf60c12a8fba5c294aa7459c60fd9f78cb640d93d","source":{"kind":"arxiv","id":"2605.06597","version":2},"attestation_state":"computed","paper":{"title":"UniSD: Towards a Unified Self-Distillation Framework for Large Language Models","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"A unified framework makes self-distillation a reliable way to adapt large language models without stronger teachers.","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"B. Aditya Prakash, Haoxin Liu, Jindong Wang, Josiah Hester, Lucheng Fu, Srijan Kumar, Yijia Xiao, Yinyi Luo, Yiqiao Jin, Yiyang Wang","submitted_at":"2026-05-07T17:22:11Z","abstract_excerpt":"Self-distillation (SD) offers a promising path for adapting large language models (LLMs) without relying on stronger external teachers. However, SD in autoregressive LLMs remains challenging because self-generated trajectories are free-form, correctness is task-dependent, and plausible rationales can still provide unstable or unreliable supervision. Existing methods mainly examine isolated design choices, leaving their effectiveness, roles, and interactions unclear. In this paper, we propose UniSD, a unified framework to systematically study self-distillation. UniSD integrates complementary me"},"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":"2605.06597","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-07T17:22:11Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"9ee1af973d04efab55c7bc8c341184da8f541824251ef46afd93b72e02342787","abstract_canon_sha256":"609fb5e8e5266b7b8008b4f1e30100c9dc1b5b20f5eb31361abcd749c2054462"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-22T02:04:41.945423Z","signature_b64":"gyF1yaXC3sPgmGSU6uyD6hY02kKytmu0mpyGTpZF+34YCgIDm6RgqMPnvAyCYwwIitOM/NliQncLk8k2SjIBCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fdc80d74b036a11468adbdcbf60c12a8fba5c294aa7459c60fd9f78cb640d93d","last_reissued_at":"2026-05-22T02:04:41.944621Z","signature_status":"signed_v1","first_computed_at":"2026-05-22T02:04:41.944621Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"UniSD: Towards a Unified Self-Distillation Framework for Large Language Models","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"A unified framework makes self-distillation a reliable way to adapt large language models without stronger teachers.","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"B. Aditya Prakash, Haoxin Liu, Jindong Wang, Josiah Hester, Lucheng Fu, Srijan Kumar, Yijia Xiao, Yinyi Luo, Yiqiao Jin, Yiyang Wang","submitted_at":"2026-05-07T17:22:11Z","abstract_excerpt":"Self-distillation (SD) offers a promising path for adapting large language models (LLMs) without relying on stronger external teachers. However, SD in autoregressive LLMs remains challenging because self-generated trajectories are free-form, correctness is task-dependent, and plausible rationales can still provide unstable or unreliable supervision. Existing methods mainly examine isolated design choices, leaving their effectiveness, roles, and interactions unclear. In this paper, we propose UniSD, a unified framework to systematically study self-distillation. UniSD integrates complementary me"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"UniSDfull, an integrated pipeline that combines complementary components, achieves the strongest overall performance, improving over the base model by +5.4 points and the strongest baseline by +2.8 points.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the listed mechanisms (multi-teacher agreement, EMA, token contrastive learning, feature matching, divergence clipping) reliably address supervision instability in free-form self-generated trajectories and that their interactions can be isolated and combined without post-hoc selection bias affecting the reported gains.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"UniSD unifies complementary self-distillation mechanisms for autoregressive LLMs and achieves up to +5.4 point gains over base models and +2.8 over baselines across six benchmarks and six models.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A unified framework makes self-distillation a reliable way to adapt large language models without stronger teachers.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"43f0993ffab088ed0c3a24d3818d0127f6c5b5d58b2ff8847fd9c854ec42b87a"},"source":{"id":"2605.06597","kind":"arxiv","version":2},"verdict":{"id":"c6a1c7da-9796-4d61-893c-c22d84397521","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-08T09:52:45.380637Z","strongest_claim":"UniSDfull, an integrated pipeline that combines complementary components, achieves the strongest overall performance, improving over the base model by +5.4 points and the strongest baseline by +2.8 points.","one_line_summary":"UniSD unifies complementary self-distillation mechanisms for autoregressive LLMs and achieves up to +5.4 point gains over base models and +2.8 over baselines across six benchmarks and six models.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the listed mechanisms (multi-teacher agreement, EMA, token contrastive learning, feature matching, divergence clipping) reliably address supervision instability in free-form self-generated trajectories and that their interactions can be isolated and combined without post-hoc selection bias affecting the reported gains.","pith_extraction_headline":"A unified framework makes self-distillation a reliable way to adapt large language models without stronger teachers."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.06597/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-20T12:22:03.785641Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-20T07:37:52.552364Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T18:01:19.364843Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T12:33:31.546986Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"162ddda1bbe6f2db18cebf495266973fcf3756ba02a57ed1f4fd0e4d62967ed0"},"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":"2605.06597","created_at":"2026-05-22T02:04:41.944753+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.06597v2","created_at":"2026-05-22T02:04:41.944753+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.06597","created_at":"2026-05-22T02:04:41.944753+00:00"},{"alias_kind":"pith_short_12","alias_value":"7XEA25FQG2QR","created_at":"2026-05-22T02:04:41.944753+00:00"},{"alias_kind":"pith_short_16","alias_value":"7XEA25FQG2QRI2FN","created_at":"2026-05-22T02:04:41.944753+00:00"},{"alias_kind":"pith_short_8","alias_value":"7XEA25FQ","created_at":"2026-05-22T02:04:41.944753+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2605.18141","citing_title":"A Brief Overview: On-Policy Self-Distillation In Large Language Models","ref_index":20,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18141","citing_title":"A Brief Overview: On-Policy Self-Distillation In Large Language Models","ref_index":20,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7XEA25FQG2QRI2FNXXF7MDASVD","json":"https://pith.science/pith/7XEA25FQG2QRI2FNXXF7MDASVD.json","graph_json":"https://pith.science/api/pith-number/7XEA25FQG2QRI2FNXXF7MDASVD/graph.json","events_json":"https://pith.science/api/pith-number/7XEA25FQG2QRI2FNXXF7MDASVD/events.json","paper":"https://pith.science/paper/7XEA25FQ"},"agent_actions":{"view_html":"https://pith.science/pith/7XEA25FQG2QRI2FNXXF7MDASVD","download_json":"https://pith.science/pith/7XEA25FQG2QRI2FNXXF7MDASVD.json","view_paper":"https://pith.science/paper/7XEA25FQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.06597&json=true","fetch_graph":"https://pith.science/api/pith-number/7XEA25FQG2QRI2FNXXF7MDASVD/graph.json","fetch_events":"https://pith.science/api/pith-number/7XEA25FQG2QRI2FNXXF7MDASVD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7XEA25FQG2QRI2FNXXF7MDASVD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7XEA25FQG2QRI2FNXXF7MDASVD/action/storage_attestation","attest_author":"https://pith.science/pith/7XEA25FQG2QRI2FNXXF7MDASVD/action/author_attestation","sign_citation":"https://pith.science/pith/7XEA25FQG2QRI2FNXXF7MDASVD/action/citation_signature","submit_replication":"https://pith.science/pith/7XEA25FQG2QRI2FNXXF7MDASVD/action/replication_record"}},"created_at":"2026-05-22T02:04:41.944753+00:00","updated_at":"2026-05-22T02:04:41.944753+00:00"}