{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:6Q7M7D7ETKKFPNRTSFRQILKV2I","short_pith_number":"pith:6Q7M7D7E","schema_version":"1.0","canonical_sha256":"f43ecf8fe49a9457b6339163042d55d2071f3a9a60b25b821c8a839e4e20c744","source":{"kind":"arxiv","id":"2606.04188","version":1},"attestation_state":"computed","paper":{"title":"Dual Advantage Fields","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.RO"],"primary_cat":"cs.LG","authors_text":"Alexander Nikulin, Alexey Zemtsov, Arip Asadulaev, Dmitry V. Dylov, Fakhri Karray, Martin Tak\\'a\\v{c}, Maxim Bobrin, Vladislav Kurenkov","submitted_at":"2026-06-02T20:15:14Z","abstract_excerpt":"Offline goal-conditioned reinforcement learning requires both long-horizon reachability estimates and local action comparisons. Dual goal representations provide value fields that capture global goal reachability, but they do not directly specify which action should be preferred at a given state. We propose Dual Advantage Fields, a policy-extraction method that turns a bilinear dual value model into a local advantage signal. Under bilinear dual parameterization, the goal embedding is the gradient of the value field with respect to the state representation. DAF learns an action-effect model tha"},"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":"2606.04188","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-02T20:15:14Z","cross_cats_sorted":["cs.AI","cs.RO"],"title_canon_sha256":"10f7f935802c8e44cd94de0976aa998f9763916146c1d6f95fc741fddfaf151b","abstract_canon_sha256":"c18cbd2ebd3b6723242161d8a818a63ed5c02729bf31abe33ea728a67d7456f2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-04T01:08:19.611053Z","signature_b64":"zoa8v3sElyQ2806tcRCS7YvndZPK2n+MPpk9ZXQQJfq9FGwv8sHOXletTtiwsvMfVsw6ru2RRu8b3OfdQiTEAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f43ecf8fe49a9457b6339163042d55d2071f3a9a60b25b821c8a839e4e20c744","last_reissued_at":"2026-06-04T01:08:19.610232Z","signature_status":"signed_v1","first_computed_at":"2026-06-04T01:08:19.610232Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Dual Advantage Fields","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.RO"],"primary_cat":"cs.LG","authors_text":"Alexander Nikulin, Alexey Zemtsov, Arip Asadulaev, Dmitry V. Dylov, Fakhri Karray, Martin Tak\\'a\\v{c}, Maxim Bobrin, Vladislav Kurenkov","submitted_at":"2026-06-02T20:15:14Z","abstract_excerpt":"Offline goal-conditioned reinforcement learning requires both long-horizon reachability estimates and local action comparisons. Dual goal representations provide value fields that capture global goal reachability, but they do not directly specify which action should be preferred at a given state. We propose Dual Advantage Fields, a policy-extraction method that turns a bilinear dual value model into a local advantage signal. Under bilinear dual parameterization, the goal embedding is the gradient of the value field with respect to the state representation. DAF learns an action-effect model tha"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.04188","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/2606.04188/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":"2606.04188","created_at":"2026-06-04T01:08:19.610359+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.04188v1","created_at":"2026-06-04T01:08:19.610359+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.04188","created_at":"2026-06-04T01:08:19.610359+00:00"},{"alias_kind":"pith_short_12","alias_value":"6Q7M7D7ETKKF","created_at":"2026-06-04T01:08:19.610359+00:00"},{"alias_kind":"pith_short_16","alias_value":"6Q7M7D7ETKKFPNRT","created_at":"2026-06-04T01:08:19.610359+00:00"},{"alias_kind":"pith_short_8","alias_value":"6Q7M7D7E","created_at":"2026-06-04T01:08:19.610359+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/6Q7M7D7ETKKFPNRTSFRQILKV2I","json":"https://pith.science/pith/6Q7M7D7ETKKFPNRTSFRQILKV2I.json","graph_json":"https://pith.science/api/pith-number/6Q7M7D7ETKKFPNRTSFRQILKV2I/graph.json","events_json":"https://pith.science/api/pith-number/6Q7M7D7ETKKFPNRTSFRQILKV2I/events.json","paper":"https://pith.science/paper/6Q7M7D7E"},"agent_actions":{"view_html":"https://pith.science/pith/6Q7M7D7ETKKFPNRTSFRQILKV2I","download_json":"https://pith.science/pith/6Q7M7D7ETKKFPNRTSFRQILKV2I.json","view_paper":"https://pith.science/paper/6Q7M7D7E","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.04188&json=true","fetch_graph":"https://pith.science/api/pith-number/6Q7M7D7ETKKFPNRTSFRQILKV2I/graph.json","fetch_events":"https://pith.science/api/pith-number/6Q7M7D7ETKKFPNRTSFRQILKV2I/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6Q7M7D7ETKKFPNRTSFRQILKV2I/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6Q7M7D7ETKKFPNRTSFRQILKV2I/action/storage_attestation","attest_author":"https://pith.science/pith/6Q7M7D7ETKKFPNRTSFRQILKV2I/action/author_attestation","sign_citation":"https://pith.science/pith/6Q7M7D7ETKKFPNRTSFRQILKV2I/action/citation_signature","submit_replication":"https://pith.science/pith/6Q7M7D7ETKKFPNRTSFRQILKV2I/action/replication_record"}},"created_at":"2026-06-04T01:08:19.610359+00:00","updated_at":"2026-06-04T01:08:19.610359+00:00"}