{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:GOCAXDHPYVDQX4URP3TBP65HYK","short_pith_number":"pith:GOCAXDHP","schema_version":"1.0","canonical_sha256":"33840b8cefc5470bf2917ee617fba7c2a1c69d5431be21f3689e6279fa053b3a","source":{"kind":"arxiv","id":"2607.04751","version":1},"attestation_state":"computed","paper":{"title":"Trust Region Policy Distillation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Li Lyna Zhang, Mao Yang, Zeke Xie, Zhengpeng Xie","submitted_at":"2026-07-06T07:43:36Z","abstract_excerpt":"Big goals are hard to achieve all at once; breaking them into small steps is wiser. We present Trust Region Policy Distillation (TOP-D), which transforms the notoriously unstable, high-variance On-Policy Distillation (OPD) into a stable training paradigm by dynamically constructing a proximal teacher. Theoretically, we establish a rigorous framework demonstrating that TOP-D inherently controls gradient variance. By providing a formal global convergence analysis alongside a monotonic improvement bound, we mathematically formalize the reliability and stability of the overall training dynamics. E"},"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":"2607.04751","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-07-06T07:43:36Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3aaa223c2d512bd14abee6b04ad6ca2f42d4f00d641a1f865d063738f49bc412","abstract_canon_sha256":"952f7cd1e2e983847c1a5a0d2b704141042c29a6768530ebc3f4d83300fd12c4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-07T02:20:01.528606Z","signature_b64":"Z2fJHtRduyEpSXKip/0PkjhAbQpF+uUYsNRD0ject4DxjF+CbAna8qjIgq3PuGFMHvMKW4cfPol3yNGAXIMzAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"33840b8cefc5470bf2917ee617fba7c2a1c69d5431be21f3689e6279fa053b3a","last_reissued_at":"2026-07-07T02:20:01.527830Z","signature_status":"signed_v1","first_computed_at":"2026-07-07T02:20:01.527830Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Trust Region Policy Distillation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Li Lyna Zhang, Mao Yang, Zeke Xie, Zhengpeng Xie","submitted_at":"2026-07-06T07:43:36Z","abstract_excerpt":"Big goals are hard to achieve all at once; breaking them into small steps is wiser. We present Trust Region Policy Distillation (TOP-D), which transforms the notoriously unstable, high-variance On-Policy Distillation (OPD) into a stable training paradigm by dynamically constructing a proximal teacher. Theoretically, we establish a rigorous framework demonstrating that TOP-D inherently controls gradient variance. By providing a formal global convergence analysis alongside a monotonic improvement bound, we mathematically formalize the reliability and stability of the overall training dynamics. E"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.04751","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/2607.04751/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":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":"2607.04751","created_at":"2026-07-07T02:20:01.527963+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.04751v1","created_at":"2026-07-07T02:20:01.527963+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.04751","created_at":"2026-07-07T02:20:01.527963+00:00"},{"alias_kind":"pith_short_12","alias_value":"GOCAXDHPYVDQ","created_at":"2026-07-07T02:20:01.527963+00:00"},{"alias_kind":"pith_short_16","alias_value":"GOCAXDHPYVDQX4UR","created_at":"2026-07-07T02:20:01.527963+00:00"},{"alias_kind":"pith_short_8","alias_value":"GOCAXDHP","created_at":"2026-07-07T02:20:01.527963+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/GOCAXDHPYVDQX4URP3TBP65HYK","json":"https://pith.science/pith/GOCAXDHPYVDQX4URP3TBP65HYK.json","graph_json":"https://pith.science/api/pith-number/GOCAXDHPYVDQX4URP3TBP65HYK/graph.json","events_json":"https://pith.science/api/pith-number/GOCAXDHPYVDQX4URP3TBP65HYK/events.json","paper":"https://pith.science/paper/GOCAXDHP"},"agent_actions":{"view_html":"https://pith.science/pith/GOCAXDHPYVDQX4URP3TBP65HYK","download_json":"https://pith.science/pith/GOCAXDHPYVDQX4URP3TBP65HYK.json","view_paper":"https://pith.science/paper/GOCAXDHP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.04751&json=true","fetch_graph":"https://pith.science/api/pith-number/GOCAXDHPYVDQX4URP3TBP65HYK/graph.json","fetch_events":"https://pith.science/api/pith-number/GOCAXDHPYVDQX4URP3TBP65HYK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GOCAXDHPYVDQX4URP3TBP65HYK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GOCAXDHPYVDQX4URP3TBP65HYK/action/storage_attestation","attest_author":"https://pith.science/pith/GOCAXDHPYVDQX4URP3TBP65HYK/action/author_attestation","sign_citation":"https://pith.science/pith/GOCAXDHPYVDQX4URP3TBP65HYK/action/citation_signature","submit_replication":"https://pith.science/pith/GOCAXDHPYVDQX4URP3TBP65HYK/action/replication_record"}},"created_at":"2026-07-07T02:20:01.527963+00:00","updated_at":"2026-07-07T02:20:01.527963+00:00"}