{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:Q5FI6A25T3UQ5HM5UDEVOYH5W2","short_pith_number":"pith:Q5FI6A25","schema_version":"1.0","canonical_sha256":"874a8f035d9ee90e9d9da0c95760fdb6998015c761c02c3c9314c3b598f86f3c","source":{"kind":"arxiv","id":"2606.29502","version":1},"attestation_state":"computed","paper":{"title":"UCOB: Learning to Utilize and Evolve Agentic Skills via Credit-Aware On-Policy Bidirectional Self-Distillation","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Chengdong Xu, Dongbin Zhao, Dong Li, Linjing Li, Qichao Zhang, Songjun Tu, Xiangyuan Lan, Yaocheng Zhang, Yiwen Ma","submitted_at":"2026-06-28T17:02:18Z","abstract_excerpt":"Skill memories can improve agentic reinforcement learning by reusing past experience as textual guidance, but retrieved skills are not oracular: they may help in one state while misleading the same policy in another. This makes the common privileged-teacher assumption fragile, namely that a skill-conditioned prompt can be treated as a fixed teacher for the no-skill prompt. We introduce UCOB, a framework for learning to utilize and evolve agentic skills via credit-aware on-policy bidirectional self-distillation. UCOB treats skill-conditioned and no-skill prompts as two on-policy context views o"},"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":"2606.29502","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-28T17:02:18Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"d2cf3df97570302a7ac17f1ad11113b5de7f58fbc980b2339f171bac6991711a","abstract_canon_sha256":"fd65251b43d2383a77db151a092ffcc5d7a911efa3c17abe7be8da6edad4038d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-30T01:18:09.446248Z","signature_b64":"NNUmIRvejF2gqd18vbu403UGt1rwf68CI6Q5eqQ+N1ySTVYgsPSJd3TS2jmb5WTpuRIkpuauuMUW8MmMblT5Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"874a8f035d9ee90e9d9da0c95760fdb6998015c761c02c3c9314c3b598f86f3c","last_reissued_at":"2026-06-30T01:18:09.445667Z","signature_status":"signed_v1","first_computed_at":"2026-06-30T01:18:09.445667Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"UCOB: Learning to Utilize and Evolve Agentic Skills via Credit-Aware On-Policy Bidirectional Self-Distillation","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Chengdong Xu, Dongbin Zhao, Dong Li, Linjing Li, Qichao Zhang, Songjun Tu, Xiangyuan Lan, Yaocheng Zhang, Yiwen Ma","submitted_at":"2026-06-28T17:02:18Z","abstract_excerpt":"Skill memories can improve agentic reinforcement learning by reusing past experience as textual guidance, but retrieved skills are not oracular: they may help in one state while misleading the same policy in another. This makes the common privileged-teacher assumption fragile, namely that a skill-conditioned prompt can be treated as a fixed teacher for the no-skill prompt. We introduce UCOB, a framework for learning to utilize and evolve agentic skills via credit-aware on-policy bidirectional self-distillation. UCOB treats skill-conditioned and no-skill prompts as two on-policy context views o"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.29502","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/2606.29502/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":"2606.29502","created_at":"2026-06-30T01:18:09.445759+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.29502v1","created_at":"2026-06-30T01:18:09.445759+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.29502","created_at":"2026-06-30T01:18:09.445759+00:00"},{"alias_kind":"pith_short_12","alias_value":"Q5FI6A25T3UQ","created_at":"2026-06-30T01:18:09.445759+00:00"},{"alias_kind":"pith_short_16","alias_value":"Q5FI6A25T3UQ5HM5","created_at":"2026-06-30T01:18:09.445759+00:00"},{"alias_kind":"pith_short_8","alias_value":"Q5FI6A25","created_at":"2026-06-30T01:18:09.445759+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/Q5FI6A25T3UQ5HM5UDEVOYH5W2","json":"https://pith.science/pith/Q5FI6A25T3UQ5HM5UDEVOYH5W2.json","graph_json":"https://pith.science/api/pith-number/Q5FI6A25T3UQ5HM5UDEVOYH5W2/graph.json","events_json":"https://pith.science/api/pith-number/Q5FI6A25T3UQ5HM5UDEVOYH5W2/events.json","paper":"https://pith.science/paper/Q5FI6A25"},"agent_actions":{"view_html":"https://pith.science/pith/Q5FI6A25T3UQ5HM5UDEVOYH5W2","download_json":"https://pith.science/pith/Q5FI6A25T3UQ5HM5UDEVOYH5W2.json","view_paper":"https://pith.science/paper/Q5FI6A25","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.29502&json=true","fetch_graph":"https://pith.science/api/pith-number/Q5FI6A25T3UQ5HM5UDEVOYH5W2/graph.json","fetch_events":"https://pith.science/api/pith-number/Q5FI6A25T3UQ5HM5UDEVOYH5W2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Q5FI6A25T3UQ5HM5UDEVOYH5W2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Q5FI6A25T3UQ5HM5UDEVOYH5W2/action/storage_attestation","attest_author":"https://pith.science/pith/Q5FI6A25T3UQ5HM5UDEVOYH5W2/action/author_attestation","sign_citation":"https://pith.science/pith/Q5FI6A25T3UQ5HM5UDEVOYH5W2/action/citation_signature","submit_replication":"https://pith.science/pith/Q5FI6A25T3UQ5HM5UDEVOYH5W2/action/replication_record"}},"created_at":"2026-06-30T01:18:09.445759+00:00","updated_at":"2026-06-30T01:18:09.445759+00:00"}