{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:VPZTCO4J4EG4HSUEEIFR2J7HZT","short_pith_number":"pith:VPZTCO4J","schema_version":"1.0","canonical_sha256":"abf3313b89e10dc3ca84220b1d27e7ccd7906413524cf5e18781579e1ba986ce","source":{"kind":"arxiv","id":"2605.13230","version":1},"attestation_state":"computed","paper":{"title":"Teacher-Guided Policy Optimization for LLM Distillation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Teacher-Guided Policy Optimization fixes uninformative feedback in reverse KL by conditioning teacher predictions on student rollouts.","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Bei Li, Chunyang Xiao, Jiahao Liu, Jingang Wang, Jingbo Zhu, Junhao Ruan, Kechen Jiao, Qifan Wang, Runsong Zhao, Tong Xiao, Xin Chen, Xinyu Liu","submitted_at":"2026-05-13T09:20:03Z","abstract_excerpt":"The convergence of reinforcement learning and imitation learning has positioned Reverse KL (RKL) as a promising paradigm for on-policy LLM distillation, aiming to unify exploration with teacher supervision. However, we identify a critical limitation: when the student and teacher distributions diverge significantly, standard RKL often fails to yield meaningful improvement due to uninformative negative feedback. To address this inefficiency, we propose Teacher-Guided Policy Optimization (TGPO), an on-policy algorithm that incorporates dense directional guidance by leveraging teacher predictions "},"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":"2605.13230","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-13T09:20:03Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"0e90ea93b8cdaae16215c07b5d8dfcd34c3814bd0bf4212ce3a2310bad428029","abstract_canon_sha256":"2c4910f7ce364db0a93c72122ee6e645a59abedf6cfd9fab5f920fd384541c66"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T02:44:49.594892Z","signature_b64":"NcPYvEzGSECok955zBjajnBefCoNVmnqeNdSnEfzbDUo16mMRPvD7aQ0j/rL4QE5aA7yPnj9AHAdJZXCP3QVCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"abf3313b89e10dc3ca84220b1d27e7ccd7906413524cf5e18781579e1ba986ce","last_reissued_at":"2026-05-18T02:44:49.594387Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T02:44:49.594387Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Teacher-Guided Policy Optimization for LLM Distillation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Teacher-Guided Policy Optimization fixes uninformative feedback in reverse KL by conditioning teacher predictions on student rollouts.","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Bei Li, Chunyang Xiao, Jiahao Liu, Jingang Wang, Jingbo Zhu, Junhao Ruan, Kechen Jiao, Qifan Wang, Runsong Zhao, Tong Xiao, Xin Chen, Xinyu Liu","submitted_at":"2026-05-13T09:20:03Z","abstract_excerpt":"The convergence of reinforcement learning and imitation learning has positioned Reverse KL (RKL) as a promising paradigm for on-policy LLM distillation, aiming to unify exploration with teacher supervision. However, we identify a critical limitation: when the student and teacher distributions diverge significantly, standard RKL often fails to yield meaningful improvement due to uninformative negative feedback. To address this inefficiency, we propose Teacher-Guided Policy Optimization (TGPO), an on-policy algorithm that incorporates dense directional guidance by leveraging teacher predictions "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"TGPO significantly outperforms standard baselines and is robust to different teachers on complex reasoning benchmarks.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That conditioning teacher predictions on the student's rollout will reliably produce informative directional guidance even when student and teacher distributions diverge substantially.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"TGPO improves on-policy LLM distillation by using teacher predictions conditioned on student rollouts to supply informative guidance when the two distributions diverge.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Teacher-Guided Policy Optimization fixes uninformative feedback in reverse KL by conditioning teacher predictions on student rollouts.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"8a2901be0c3cc032ae152b0d89721a4915a10996bb763507ee2b68437eb85823"},"source":{"id":"2605.13230","kind":"arxiv","version":1},"verdict":{"id":"34e18f50-4e1d-47e3-b594-ffb42a21d78a","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T20:08:08.482193Z","strongest_claim":"TGPO significantly outperforms standard baselines and is robust to different teachers on complex reasoning benchmarks.","one_line_summary":"TGPO improves on-policy LLM distillation by using teacher predictions conditioned on student rollouts to supply informative guidance when the two distributions diverge.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That conditioning teacher predictions on the student's rollout will reliably produce informative directional guidance even when student and teacher distributions diverge substantially.","pith_extraction_headline":"Teacher-Guided Policy Optimization fixes uninformative feedback in reverse KL by conditioning teacher predictions on student rollouts."},"references":{"count":23,"sample":[{"doi":"","year":null,"title":"SFT or RL? An Early Investigation into Training R1-Like Reasoning Large Vision-Language Models","work_id":"a521360c-8673-4d0d-a3a3-6eb9f7a71b90","ref_index":1,"cited_arxiv_id":"2504.11468","is_internal_anchor":true},{"doi":"","year":null,"title":"SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training","work_id":"258dd934-025c-47f5-b4f6-5a0c1c338cc6","ref_index":2,"cited_arxiv_id":"2501.17161","is_internal_anchor":true},{"doi":"","year":null,"title":"Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge","work_id":"28ea1282-d657-4c61-a83c-f1249be6d6b1","ref_index":3,"cited_arxiv_id":"1803.05457","is_internal_anchor":true},{"doi":"","year":null,"title":"Process Reinforcement through Implicit Rewards","work_id":"c31a2126-86f9-44f3-91f3-208d0fc1463a","ref_index":4,"cited_arxiv_id":"2502.01456","is_internal_anchor":true},{"doi":"","year":null,"title":"MiniLLM: On-Policy Distillation of Large Language Models","work_id":"16edb291-dd18-41c5-8486-c6c715ec5311","ref_index":5,"cited_arxiv_id":"2306.08543","is_internal_anchor":true}],"resolved_work":23,"snapshot_sha256":"6b4db49c37be5a137b76f46ca181a3ed485126ffe3b526aef6223a9f420410de","internal_anchors":18},"formal_canon":{"evidence_count":2,"snapshot_sha256":"38b4428d457f6357cc5dd19881c66c582aecf2de28b6117742a1e70fd9a027cd"},"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":"2605.13230","created_at":"2026-05-18T02:44:49.594474+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.13230v1","created_at":"2026-05-18T02:44:49.594474+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.13230","created_at":"2026-05-18T02:44:49.594474+00:00"},{"alias_kind":"pith_short_12","alias_value":"VPZTCO4J4EG4","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"VPZTCO4J4EG4HSUE","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"VPZTCO4J","created_at":"2026-05-18T12:33:37.589309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VPZTCO4J4EG4HSUEEIFR2J7HZT","json":"https://pith.science/pith/VPZTCO4J4EG4HSUEEIFR2J7HZT.json","graph_json":"https://pith.science/api/pith-number/VPZTCO4J4EG4HSUEEIFR2J7HZT/graph.json","events_json":"https://pith.science/api/pith-number/VPZTCO4J4EG4HSUEEIFR2J7HZT/events.json","paper":"https://pith.science/paper/VPZTCO4J"},"agent_actions":{"view_html":"https://pith.science/pith/VPZTCO4J4EG4HSUEEIFR2J7HZT","download_json":"https://pith.science/pith/VPZTCO4J4EG4HSUEEIFR2J7HZT.json","view_paper":"https://pith.science/paper/VPZTCO4J","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.13230&json=true","fetch_graph":"https://pith.science/api/pith-number/VPZTCO4J4EG4HSUEEIFR2J7HZT/graph.json","fetch_events":"https://pith.science/api/pith-number/VPZTCO4J4EG4HSUEEIFR2J7HZT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VPZTCO4J4EG4HSUEEIFR2J7HZT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VPZTCO4J4EG4HSUEEIFR2J7HZT/action/storage_attestation","attest_author":"https://pith.science/pith/VPZTCO4J4EG4HSUEEIFR2J7HZT/action/author_attestation","sign_citation":"https://pith.science/pith/VPZTCO4J4EG4HSUEEIFR2J7HZT/action/citation_signature","submit_replication":"https://pith.science/pith/VPZTCO4J4EG4HSUEEIFR2J7HZT/action/replication_record"}},"created_at":"2026-05-18T02:44:49.594474+00:00","updated_at":"2026-05-18T02:44:49.594474+00:00"}