{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:LJKOVZP6MDUJLQL2GSEOCIX653","short_pith_number":"pith:LJKOVZP6","schema_version":"1.0","canonical_sha256":"5a54eae5fe60e895c17a3488e122feeeefcc17c7a10d2419fc1cac4cbfb39a26","source":{"kind":"arxiv","id":"2605.29584","version":1},"attestation_state":"computed","paper":{"title":"GAPD: Gold-Action Policy Distillation for Agentic Reinforcement Learning in Knowledge Base Question Answering","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bowen Song, Jianan Xie, Liang Wang, Qiang Liu, Shu Wu, Weiqiang Wang, Xin Sun, Zhongqi Chen, Zilei Wang","submitted_at":"2026-05-28T08:28:10Z","abstract_excerpt":"Reinforcement learning (RL) is a natural fit for agentic knowledge base question answering (KBQA), where a model must issue executable actions, observe knowledge-base feedback, and eventually return an answer. However, current RL-based KBQA systems mainly optimize sparse rewards from the final answer, leaving intermediate action errors weakly supervised. This is especially limiting for logical-form annotated KBQA benchmarks: gold logical forms can be converted into executable action sequences, but existing pipelines use them mainly for warm-start data construction rather than for on-policy RL "},"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":"2605.29584","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-28T08:28:10Z","cross_cats_sorted":[],"title_canon_sha256":"37cffb8ee7a69d91136785b25f28fdf5ea9e094b4b9cd0fe4d22cc0e16bc2f5f","abstract_canon_sha256":"88d8671d2ff0ba888fddea4e87175a8ff8e498c5e4dd5fc09c594995eb27b140"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T01:05:49.077677Z","signature_b64":"EUvC9UnCkbzG7zUj2T8bm8x7AQBSY9zMwEONvk2Y3gW8OvXjuElmMQiqERsb9WDYV6Xg8dPBYbsdIBCB1/m3Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5a54eae5fe60e895c17a3488e122feeeefcc17c7a10d2419fc1cac4cbfb39a26","last_reissued_at":"2026-05-29T01:05:49.076730Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T01:05:49.076730Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"GAPD: Gold-Action Policy Distillation for Agentic Reinforcement Learning in Knowledge Base Question Answering","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bowen Song, Jianan Xie, Liang Wang, Qiang Liu, Shu Wu, Weiqiang Wang, Xin Sun, Zhongqi Chen, Zilei Wang","submitted_at":"2026-05-28T08:28:10Z","abstract_excerpt":"Reinforcement learning (RL) is a natural fit for agentic knowledge base question answering (KBQA), where a model must issue executable actions, observe knowledge-base feedback, and eventually return an answer. However, current RL-based KBQA systems mainly optimize sparse rewards from the final answer, leaving intermediate action errors weakly supervised. This is especially limiting for logical-form annotated KBQA benchmarks: gold logical forms can be converted into executable action sequences, but existing pipelines use them mainly for warm-start data construction rather than for on-policy RL "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.29584","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/2605.29584/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":"2605.29584","created_at":"2026-05-29T01:05:49.076854+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.29584v1","created_at":"2026-05-29T01:05:49.076854+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.29584","created_at":"2026-05-29T01:05:49.076854+00:00"},{"alias_kind":"pith_short_12","alias_value":"LJKOVZP6MDUJ","created_at":"2026-05-29T01:05:49.076854+00:00"},{"alias_kind":"pith_short_16","alias_value":"LJKOVZP6MDUJLQL2","created_at":"2026-05-29T01:05:49.076854+00:00"},{"alias_kind":"pith_short_8","alias_value":"LJKOVZP6","created_at":"2026-05-29T01:05:49.076854+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/LJKOVZP6MDUJLQL2GSEOCIX653","json":"https://pith.science/pith/LJKOVZP6MDUJLQL2GSEOCIX653.json","graph_json":"https://pith.science/api/pith-number/LJKOVZP6MDUJLQL2GSEOCIX653/graph.json","events_json":"https://pith.science/api/pith-number/LJKOVZP6MDUJLQL2GSEOCIX653/events.json","paper":"https://pith.science/paper/LJKOVZP6"},"agent_actions":{"view_html":"https://pith.science/pith/LJKOVZP6MDUJLQL2GSEOCIX653","download_json":"https://pith.science/pith/LJKOVZP6MDUJLQL2GSEOCIX653.json","view_paper":"https://pith.science/paper/LJKOVZP6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.29584&json=true","fetch_graph":"https://pith.science/api/pith-number/LJKOVZP6MDUJLQL2GSEOCIX653/graph.json","fetch_events":"https://pith.science/api/pith-number/LJKOVZP6MDUJLQL2GSEOCIX653/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LJKOVZP6MDUJLQL2GSEOCIX653/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LJKOVZP6MDUJLQL2GSEOCIX653/action/storage_attestation","attest_author":"https://pith.science/pith/LJKOVZP6MDUJLQL2GSEOCIX653/action/author_attestation","sign_citation":"https://pith.science/pith/LJKOVZP6MDUJLQL2GSEOCIX653/action/citation_signature","submit_replication":"https://pith.science/pith/LJKOVZP6MDUJLQL2GSEOCIX653/action/replication_record"}},"created_at":"2026-05-29T01:05:49.076854+00:00","updated_at":"2026-05-29T01:05:49.076854+00:00"}