{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:CBEEOBLGPICEHEA2Z4XLNAIJXK","short_pith_number":"pith:CBEEOBLG","schema_version":"1.0","canonical_sha256":"10484705667a0443901acf2eb68109ba9d838e9da2fdc74055d9ff3350a18930","source":{"kind":"arxiv","id":"2605.15155","version":1},"attestation_state":"computed","paper":{"title":"Self-Distilled Agentic Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Jinyang Wu, Jun Xiao, Qi Gu, Weiming Lu, Xunliang Cai, Yongliang Shen, Yueting Zhuang, Zhengxi Lu, Zhiyuan Yao, Zhuowen Han, Zi-Han Wang","submitted_at":"2026-05-14T17:51:26Z","abstract_excerpt":"Reinforcement learning (RL) has emerged as a central paradigm for post-training LLM agents, yet its trajectory-level reward signal provides only coarse supervision for long-horizon interaction. On-Policy Self-Distillation (OPSD) complements RL by introducing dense token-level guidance from a teacher branch augmented with privileged context. However, transferring OPSD to multi-turn agents proves problematic: compounding multi-turn instability destabilizes supervision, while skill-conditioned privileged guidance requires asymmetric treatment for negative teacher rejections may arise from imperfe"},"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":"2605.15155","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-14T17:51:26Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"60abd1093f1e5ad88b032b32f4d827f812f2b30903d70e62abea598a1d78bda8","abstract_canon_sha256":"a9184d7c6d59ef3af229b21f02699e96db38448ff469e64bc9b57a58fa5de679"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:38:53.978756Z","signature_b64":"H7zF0MJ4WpHwJWCwO/66bYG7144JPDEg4pTYpr9QzeATohBggB435wvpmRjdTBUb79vKWjH1gqizdd3SJkonDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"10484705667a0443901acf2eb68109ba9d838e9da2fdc74055d9ff3350a18930","last_reissued_at":"2026-05-17T23:38:53.978198Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:38:53.978198Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Self-Distilled Agentic Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Jinyang Wu, Jun Xiao, Qi Gu, Weiming Lu, Xunliang Cai, Yongliang Shen, Yueting Zhuang, Zhengxi Lu, Zhiyuan Yao, Zhuowen Han, Zi-Han Wang","submitted_at":"2026-05-14T17:51:26Z","abstract_excerpt":"Reinforcement learning (RL) has emerged as a central paradigm for post-training LLM agents, yet its trajectory-level reward signal provides only coarse supervision for long-horizon interaction. On-Policy Self-Distillation (OPSD) complements RL by introducing dense token-level guidance from a teacher branch augmented with privileged context. However, transferring OPSD to multi-turn agents proves problematic: compounding multi-turn instability destabilizes supervision, while skill-conditioned privileged guidance requires asymmetric treatment for negative teacher rejections may arise from imperfe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.15155","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":""},"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":"2605.15155","created_at":"2026-05-17T23:38:53.978285+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.15155v1","created_at":"2026-05-17T23:38:53.978285+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.15155","created_at":"2026-05-17T23:38:53.978285+00:00"},{"alias_kind":"pith_short_12","alias_value":"CBEEOBLGPICE","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"CBEEOBLGPICEHEA2","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"CBEEOBLG","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":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CBEEOBLGPICEHEA2Z4XLNAIJXK","json":"https://pith.science/pith/CBEEOBLGPICEHEA2Z4XLNAIJXK.json","graph_json":"https://pith.science/api/pith-number/CBEEOBLGPICEHEA2Z4XLNAIJXK/graph.json","events_json":"https://pith.science/api/pith-number/CBEEOBLGPICEHEA2Z4XLNAIJXK/events.json","paper":"https://pith.science/paper/CBEEOBLG"},"agent_actions":{"view_html":"https://pith.science/pith/CBEEOBLGPICEHEA2Z4XLNAIJXK","download_json":"https://pith.science/pith/CBEEOBLGPICEHEA2Z4XLNAIJXK.json","view_paper":"https://pith.science/paper/CBEEOBLG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.15155&json=true","fetch_graph":"https://pith.science/api/pith-number/CBEEOBLGPICEHEA2Z4XLNAIJXK/graph.json","fetch_events":"https://pith.science/api/pith-number/CBEEOBLGPICEHEA2Z4XLNAIJXK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CBEEOBLGPICEHEA2Z4XLNAIJXK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CBEEOBLGPICEHEA2Z4XLNAIJXK/action/storage_attestation","attest_author":"https://pith.science/pith/CBEEOBLGPICEHEA2Z4XLNAIJXK/action/author_attestation","sign_citation":"https://pith.science/pith/CBEEOBLGPICEHEA2Z4XLNAIJXK/action/citation_signature","submit_replication":"https://pith.science/pith/CBEEOBLGPICEHEA2Z4XLNAIJXK/action/replication_record"}},"created_at":"2026-05-17T23:38:53.978285+00:00","updated_at":"2026-05-17T23:38:53.978285+00:00"}