{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:NGSJXOMNMWSRYBOIONSOZLABM2","short_pith_number":"pith:NGSJXOMN","canonical_record":{"source":{"id":"2606.00350","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-29T20:42:30Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"159730241530d82c11d55bf15f4fcc8135ec2d0f651bd16920c344d30fbc1ccb","abstract_canon_sha256":"f5232ce9c71efcd465bd8fbd15e62221b367fda9094df7531fce86a678434235"},"schema_version":"1.0"},"canonical_sha256":"69a49bb98d65a51c05c87364ecac0166911b9f5d7392a1f05cefb9af1ee8984d","source":{"kind":"arxiv","id":"2606.00350","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.00350","created_at":"2026-06-02T01:03:52Z"},{"alias_kind":"arxiv_version","alias_value":"2606.00350v1","created_at":"2026-06-02T01:03:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.00350","created_at":"2026-06-02T01:03:52Z"},{"alias_kind":"pith_short_12","alias_value":"NGSJXOMNMWSR","created_at":"2026-06-02T01:03:52Z"},{"alias_kind":"pith_short_16","alias_value":"NGSJXOMNMWSRYBOI","created_at":"2026-06-02T01:03:52Z"},{"alias_kind":"pith_short_8","alias_value":"NGSJXOMN","created_at":"2026-06-02T01:03:52Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:NGSJXOMNMWSRYBOIONSOZLABM2","target":"record","payload":{"canonical_record":{"source":{"id":"2606.00350","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-29T20:42:30Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"159730241530d82c11d55bf15f4fcc8135ec2d0f651bd16920c344d30fbc1ccb","abstract_canon_sha256":"f5232ce9c71efcd465bd8fbd15e62221b367fda9094df7531fce86a678434235"},"schema_version":"1.0"},"canonical_sha256":"69a49bb98d65a51c05c87364ecac0166911b9f5d7392a1f05cefb9af1ee8984d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T01:03:52.344799Z","signature_b64":"SeEPOPNwQIASy4Cc7eItyHAIASLZ9uaTvTfYKgC5H+DzzPhYWTWewQUDafXKtuMZCag83JWtEpAcMc1FKIZcBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"69a49bb98d65a51c05c87364ecac0166911b9f5d7392a1f05cefb9af1ee8984d","last_reissued_at":"2026-06-02T01:03:52.344380Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T01:03:52.344380Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.00350","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-02T01:03:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DDIHn2yrlZ5lTqM3DEUGzIX+amR7ptxeT08eYr4hrKT1R6waJuapnnGYMzRtUqFrsona6AIwZROZVGAsY9lTCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-03T20:54:54.548587Z"},"content_sha256":"5659441bb812d06e969b7d2d643b04d1f7e00bb3f145505b1c9d15e914eff7dc","schema_version":"1.0","event_id":"sha256:5659441bb812d06e969b7d2d643b04d1f7e00bb3f145505b1c9d15e914eff7dc"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:NGSJXOMNMWSRYBOIONSOZLABM2","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Drift Q-Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Amin Abyaneh, Anas Houssaini, David Meger, Hsiu-Chin Lin, Mohamad H. Danesh, Scott Fujimoto","submitted_at":"2026-05-29T20:42:30Z","abstract_excerpt":"Offline reinforcement learning requires improving a policy from fixed data while avoiding out-of-distribution actions with unreliable value estimates. Diffusion and flow policies handle this trade-off by modeling the behavior distribution to regularize the RL objective, but they require iterative denoising, solver integrations, and in more efficient variants, distillation or other approximations at inference. We propose DriftQL, which combines a drift-based behavioral regularizer with critic-driven policy improvement. The value signal biases the policy toward high-value regions of the data sup"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.00350","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/2606.00350/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-02T01:03:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SFvJRtgwdTO9qVIbwROq9mVuwkBfaM3Mht81Tp61mQT8gi8tl2sAUHA6gd8S045u6j0FHmtioHSBodCJc5IvBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-03T20:54:54.548962Z"},"content_sha256":"5a441ab916298cae3a4864325562a2d46dbf224a07c9045a74846ba7b18d4c28","schema_version":"1.0","event_id":"sha256:5a441ab916298cae3a4864325562a2d46dbf224a07c9045a74846ba7b18d4c28"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NGSJXOMNMWSRYBOIONSOZLABM2/bundle.json","state_url":"https://pith.science/pith/NGSJXOMNMWSRYBOIONSOZLABM2/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NGSJXOMNMWSRYBOIONSOZLABM2/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-06-03T20:54:54Z","links":{"resolver":"https://pith.science/pith/NGSJXOMNMWSRYBOIONSOZLABM2","bundle":"https://pith.science/pith/NGSJXOMNMWSRYBOIONSOZLABM2/bundle.json","state":"https://pith.science/pith/NGSJXOMNMWSRYBOIONSOZLABM2/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NGSJXOMNMWSRYBOIONSOZLABM2/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:NGSJXOMNMWSRYBOIONSOZLABM2","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"f5232ce9c71efcd465bd8fbd15e62221b367fda9094df7531fce86a678434235","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-29T20:42:30Z","title_canon_sha256":"159730241530d82c11d55bf15f4fcc8135ec2d0f651bd16920c344d30fbc1ccb"},"schema_version":"1.0","source":{"id":"2606.00350","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.00350","created_at":"2026-06-02T01:03:52Z"},{"alias_kind":"arxiv_version","alias_value":"2606.00350v1","created_at":"2026-06-02T01:03:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.00350","created_at":"2026-06-02T01:03:52Z"},{"alias_kind":"pith_short_12","alias_value":"NGSJXOMNMWSR","created_at":"2026-06-02T01:03:52Z"},{"alias_kind":"pith_short_16","alias_value":"NGSJXOMNMWSRYBOI","created_at":"2026-06-02T01:03:52Z"},{"alias_kind":"pith_short_8","alias_value":"NGSJXOMN","created_at":"2026-06-02T01:03:52Z"}],"graph_snapshots":[{"event_id":"sha256:5a441ab916298cae3a4864325562a2d46dbf224a07c9045a74846ba7b18d4c28","target":"graph","created_at":"2026-06-02T01:03:52Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2606.00350/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Offline reinforcement learning requires improving a policy from fixed data while avoiding out-of-distribution actions with unreliable value estimates. Diffusion and flow policies handle this trade-off by modeling the behavior distribution to regularize the RL objective, but they require iterative denoising, solver integrations, and in more efficient variants, distillation or other approximations at inference. We propose DriftQL, which combines a drift-based behavioral regularizer with critic-driven policy improvement. The value signal biases the policy toward high-value regions of the data sup","authors_text":"Amin Abyaneh, Anas Houssaini, David Meger, Hsiu-Chin Lin, Mohamad H. Danesh, Scott Fujimoto","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-29T20:42:30Z","title":"Drift Q-Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.00350","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:5659441bb812d06e969b7d2d643b04d1f7e00bb3f145505b1c9d15e914eff7dc","target":"record","created_at":"2026-06-02T01:03:52Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"f5232ce9c71efcd465bd8fbd15e62221b367fda9094df7531fce86a678434235","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-29T20:42:30Z","title_canon_sha256":"159730241530d82c11d55bf15f4fcc8135ec2d0f651bd16920c344d30fbc1ccb"},"schema_version":"1.0","source":{"id":"2606.00350","kind":"arxiv","version":1}},"canonical_sha256":"69a49bb98d65a51c05c87364ecac0166911b9f5d7392a1f05cefb9af1ee8984d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"69a49bb98d65a51c05c87364ecac0166911b9f5d7392a1f05cefb9af1ee8984d","first_computed_at":"2026-06-02T01:03:52.344380Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T01:03:52.344380Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"SeEPOPNwQIASy4Cc7eItyHAIASLZ9uaTvTfYKgC5H+DzzPhYWTWewQUDafXKtuMZCag83JWtEpAcMc1FKIZcBA==","signature_status":"signed_v1","signed_at":"2026-06-02T01:03:52.344799Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.00350","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5659441bb812d06e969b7d2d643b04d1f7e00bb3f145505b1c9d15e914eff7dc","sha256:5a441ab916298cae3a4864325562a2d46dbf224a07c9045a74846ba7b18d4c28"],"state_sha256":"e9f82fbb23981efb61e057e0cd53ced7b394675ec772f8a1b17d2528d6a7bdb6"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3ZFeUabhufzOjKjctVUylQ81ewEEWSr9Ge2E0dbFPi4bCTkBqaj/qfHoGyG9V7OWacbQyOr0FRVGp2YmVTq2Ag==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-03T20:54:54.551070Z","bundle_sha256":"4993f2f6dee786cc5ee7c0837a299574a82d0055723e44d299c44fb102438f88"}}