{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:Z75IBECO6DCUAOTKAEKSRSIY3V","short_pith_number":"pith:Z75IBECO","schema_version":"1.0","canonical_sha256":"cffa80904ef0c5403a6a011528c918dd77e0ff247980c58bd51c7197566f926d","source":{"kind":"arxiv","id":"2608.07371","version":1},"attestation_state":"computed","paper":{"title":"Trajectory-Relative Hindsight Distillation for Agentic Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Haoyu Zheng, Qing Wang, Wenqiao Zhang, Yun Zhu","submitted_at":"2026-08-07T16:12:58Z","abstract_excerpt":"Recent agentic reinforcement learning methods use hindsight to complement sparse outcome rewards. However, a completed rollout can yield many such signals, leaving their appropriate allocation across turns unclear. We introduce TRIAL, a trajectory-relative hindsight distillation framework with a unified turn-aligned scoring protocol. For each decision turn, TRIAL extracts an outcome view of that decision's realized consequence and evaluates the same response under ordinary and hindsight-conditioned contexts. The signed log-probability gap determines the direction and local strength of token-le"},"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":"2608.07371","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-08-07T16:12:58Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"2269aef3ee3f52ebf367538943bc996c5ad814e31bb3949a17c78894cb0b0574","abstract_canon_sha256":"2f025ff2d1c04d5f22ab15ff66f7052c0c026b5c66fffd66c0637b5f2a3f9172"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-10T01:15:10.861727Z","signature_b64":"pkAt5RyXRnW7Hjxq9ua7/u7vrJHD4ynwS5COrommWOAdhZrPtrS29qoqaZnG7t1jvXqxoNXMrw3Etmo8Q8NtAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cffa80904ef0c5403a6a011528c918dd77e0ff247980c58bd51c7197566f926d","last_reissued_at":"2026-08-10T01:15:10.859468Z","signature_status":"signed_v1","first_computed_at":"2026-08-10T01:15:10.859468Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Trajectory-Relative Hindsight Distillation for Agentic Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Haoyu Zheng, Qing Wang, Wenqiao Zhang, Yun Zhu","submitted_at":"2026-08-07T16:12:58Z","abstract_excerpt":"Recent agentic reinforcement learning methods use hindsight to complement sparse outcome rewards. However, a completed rollout can yield many such signals, leaving their appropriate allocation across turns unclear. We introduce TRIAL, a trajectory-relative hindsight distillation framework with a unified turn-aligned scoring protocol. For each decision turn, TRIAL extracts an outcome view of that decision's realized consequence and evaluates the same response under ordinary and hindsight-conditioned contexts. The signed log-probability gap determines the direction and local strength of token-le"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.07371","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/2608.07371/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":"2608.07371","created_at":"2026-08-10T01:15:10.860323+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.07371v1","created_at":"2026-08-10T01:15:10.860323+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.07371","created_at":"2026-08-10T01:15:10.860323+00:00"},{"alias_kind":"pith_short_12","alias_value":"Z75IBECO6DCU","created_at":"2026-08-10T01:15:10.860323+00:00"},{"alias_kind":"pith_short_16","alias_value":"Z75IBECO6DCUAOTK","created_at":"2026-08-10T01:15:10.860323+00:00"},{"alias_kind":"pith_short_8","alias_value":"Z75IBECO","created_at":"2026-08-10T01:15:10.860323+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/Z75IBECO6DCUAOTKAEKSRSIY3V","json":"https://pith.science/pith/Z75IBECO6DCUAOTKAEKSRSIY3V.json","graph_json":"https://pith.science/api/pith-number/Z75IBECO6DCUAOTKAEKSRSIY3V/graph.json","events_json":"https://pith.science/api/pith-number/Z75IBECO6DCUAOTKAEKSRSIY3V/events.json","paper":"https://pith.science/paper/Z75IBECO"},"agent_actions":{"view_html":"https://pith.science/pith/Z75IBECO6DCUAOTKAEKSRSIY3V","download_json":"https://pith.science/pith/Z75IBECO6DCUAOTKAEKSRSIY3V.json","view_paper":"https://pith.science/paper/Z75IBECO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.07371&json=true","fetch_graph":"https://pith.science/api/pith-number/Z75IBECO6DCUAOTKAEKSRSIY3V/graph.json","fetch_events":"https://pith.science/api/pith-number/Z75IBECO6DCUAOTKAEKSRSIY3V/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Z75IBECO6DCUAOTKAEKSRSIY3V/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Z75IBECO6DCUAOTKAEKSRSIY3V/action/storage_attestation","attest_author":"https://pith.science/pith/Z75IBECO6DCUAOTKAEKSRSIY3V/action/author_attestation","sign_citation":"https://pith.science/pith/Z75IBECO6DCUAOTKAEKSRSIY3V/action/citation_signature","submit_replication":"https://pith.science/pith/Z75IBECO6DCUAOTKAEKSRSIY3V/action/replication_record"}},"created_at":"2026-08-10T01:15:10.860323+00:00","updated_at":"2026-08-10T01:15:10.860323+00:00"}