{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:N76UT4SSB6XFSD2LLI4QWBM2YT","short_pith_number":"pith:N76UT4SS","schema_version":"1.0","canonical_sha256":"6ffd49f2520fae590f4b5a390b059ac4ccbc80d1d4667f8eb6ca535d4d841dbe","source":{"kind":"arxiv","id":"2509.23071","version":2},"attestation_state":"computed","paper":{"title":"From Evidence to Trajectory: Abductive Reasoning Path Synthesis for Retrieval-Augmented Generation Agents Development","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ho-fung Leung, Irwin King, Jinhu Qi, Liheng Ma, Minghao Zhao, Muzhi Li, Xinyu Wang, Yifan Li, Yihong Wu, Yingxue Zhang, Zhenghan Tai, Zixing Song","submitted_at":"2025-09-27T02:53:09Z","abstract_excerpt":"Retrieval-augmented generation (RAG) agent development is hindered by the lack of executable ground-truth agent-environment interaction trajectories. Existing datasets provide questions, answers, and evidence, but lack fine-grained supervision for retriever invocation, dynamic planning, and stepwise decision-making. Reinforcement learning offers a potential solution, but often suffers from sparse rewards and cold-start failures when base large language models (LLMs) lack sufficient reasoning capability. Meanwhile, existing data synthesis methods mainly generate post-hoc rationales rather than "},"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":"2509.23071","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-27T02:53:09Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"466e68a1c0e1d652d72340881c1315e8669946b039d7d2f4d1a18993a39316d8","abstract_canon_sha256":"9653dfea3d8c329c0603c5db719c49543ff3afae003a5b8fabea36eaaca0e6eb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-21T02:21:22.002248Z","signature_b64":"JbZU3OTceaR+ImvUbLGdwwIvEQYDNqiWzhwL8MwqELyOA7+jx40es2c4DEA6G9eS00XXk61k7UFtTJ+SybQqBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6ffd49f2520fae590f4b5a390b059ac4ccbc80d1d4667f8eb6ca535d4d841dbe","last_reissued_at":"2026-07-21T02:21:22.001420Z","signature_status":"signed_v1","first_computed_at":"2026-07-21T02:21:22.001420Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"From Evidence to Trajectory: Abductive Reasoning Path Synthesis for Retrieval-Augmented Generation Agents Development","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ho-fung Leung, Irwin King, Jinhu Qi, Liheng Ma, Minghao Zhao, Muzhi Li, Xinyu Wang, Yifan Li, Yihong Wu, Yingxue Zhang, Zhenghan Tai, Zixing Song","submitted_at":"2025-09-27T02:53:09Z","abstract_excerpt":"Retrieval-augmented generation (RAG) agent development is hindered by the lack of executable ground-truth agent-environment interaction trajectories. Existing datasets provide questions, answers, and evidence, but lack fine-grained supervision for retriever invocation, dynamic planning, and stepwise decision-making. Reinforcement learning offers a potential solution, but often suffers from sparse rewards and cold-start failures when base large language models (LLMs) lack sufficient reasoning capability. Meanwhile, existing data synthesis methods mainly generate post-hoc rationales rather than "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.23071","kind":"arxiv","version":2},"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/2509.23071/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":"2509.23071","created_at":"2026-07-21T02:21:22.001842+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.23071v2","created_at":"2026-07-21T02:21:22.001842+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.23071","created_at":"2026-07-21T02:21:22.001842+00:00"},{"alias_kind":"pith_short_12","alias_value":"N76UT4SSB6XF","created_at":"2026-07-21T02:21:22.001842+00:00"},{"alias_kind":"pith_short_16","alias_value":"N76UT4SSB6XFSD2L","created_at":"2026-07-21T02:21:22.001842+00:00"},{"alias_kind":"pith_short_8","alias_value":"N76UT4SS","created_at":"2026-07-21T02:21:22.001842+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2602.14469","citing_title":"Measuring and Mitigating Post-hoc Rationalization in Reverse Chain-of-Thought Generation","ref_index":3,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/N76UT4SSB6XFSD2LLI4QWBM2YT","json":"https://pith.science/pith/N76UT4SSB6XFSD2LLI4QWBM2YT.json","graph_json":"https://pith.science/api/pith-number/N76UT4SSB6XFSD2LLI4QWBM2YT/graph.json","events_json":"https://pith.science/api/pith-number/N76UT4SSB6XFSD2LLI4QWBM2YT/events.json","paper":"https://pith.science/paper/N76UT4SS"},"agent_actions":{"view_html":"https://pith.science/pith/N76UT4SSB6XFSD2LLI4QWBM2YT","download_json":"https://pith.science/pith/N76UT4SSB6XFSD2LLI4QWBM2YT.json","view_paper":"https://pith.science/paper/N76UT4SS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.23071&json=true","fetch_graph":"https://pith.science/api/pith-number/N76UT4SSB6XFSD2LLI4QWBM2YT/graph.json","fetch_events":"https://pith.science/api/pith-number/N76UT4SSB6XFSD2LLI4QWBM2YT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/N76UT4SSB6XFSD2LLI4QWBM2YT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/N76UT4SSB6XFSD2LLI4QWBM2YT/action/storage_attestation","attest_author":"https://pith.science/pith/N76UT4SSB6XFSD2LLI4QWBM2YT/action/author_attestation","sign_citation":"https://pith.science/pith/N76UT4SSB6XFSD2LLI4QWBM2YT/action/citation_signature","submit_replication":"https://pith.science/pith/N76UT4SSB6XFSD2LLI4QWBM2YT/action/replication_record"}},"created_at":"2026-07-21T02:21:22.001842+00:00","updated_at":"2026-07-21T02:21:22.001842+00:00"}