{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:IOVTADSAIQNRQIZOM5ICML5RTE","short_pith_number":"pith:IOVTADSA","canonical_record":{"source":{"id":"2604.17415","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-04-19T12:47:52Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"a9b9c9999d2b319856cd98f141b6c5316e3159c1ff90fcf1fc01e0069613e15f","abstract_canon_sha256":"965112703ecf90a21ea7fa74bb4649f763107130cbece4ac12a9ddcc00b3251d"},"schema_version":"1.0"},"canonical_sha256":"43ab300e40441b18232e6750262fb19903d11d9b06c5390de10b5769fee83e8f","source":{"kind":"arxiv","id":"2604.17415","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.17415","created_at":"2026-06-02T01:03:47Z"},{"alias_kind":"arxiv_version","alias_value":"2604.17415v3","created_at":"2026-06-02T01:03:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.17415","created_at":"2026-06-02T01:03:47Z"},{"alias_kind":"pith_short_12","alias_value":"IOVTADSAIQNR","created_at":"2026-06-02T01:03:47Z"},{"alias_kind":"pith_short_16","alias_value":"IOVTADSAIQNRQIZO","created_at":"2026-06-02T01:03:47Z"},{"alias_kind":"pith_short_8","alias_value":"IOVTADSA","created_at":"2026-06-02T01:03:47Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:IOVTADSAIQNRQIZOM5ICML5RTE","target":"record","payload":{"canonical_record":{"source":{"id":"2604.17415","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-04-19T12:47:52Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"a9b9c9999d2b319856cd98f141b6c5316e3159c1ff90fcf1fc01e0069613e15f","abstract_canon_sha256":"965112703ecf90a21ea7fa74bb4649f763107130cbece4ac12a9ddcc00b3251d"},"schema_version":"1.0"},"canonical_sha256":"43ab300e40441b18232e6750262fb19903d11d9b06c5390de10b5769fee83e8f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T01:03:47.240129Z","signature_b64":"stWrP+g57acQKZrSwwom9zJ9sJUEl99Du3ZocIUXw5N3CFk70zaXftmv206rNfz+7dULLWpTV2bfZgmKU2VGAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"43ab300e40441b18232e6750262fb19903d11d9b06c5390de10b5769fee83e8f","last_reissued_at":"2026-06-02T01:03:47.239586Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T01:03:47.239586Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2604.17415","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-02T01:03:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"BeT7jHD1Wi2SSWl8OSgyecuf2cTm777vE+E86QnnrK3setD6jidZWyP0f7B7u6nP5akTh6p4P9gzKk2YX9pEDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-06T12:55:07.164002Z"},"content_sha256":"008857f3884c6345f9ea4cb6142d6061174e62b4d0a095f89e429627ac8101b1","schema_version":"1.0","event_id":"sha256:008857f3884c6345f9ea4cb6142d6061174e62b4d0a095f89e429627ac8101b1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:IOVTADSAIQNRQIZOM5ICML5RTE","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Reward Score Matching: Unifying Reward-based Fine-tuning for Flow and Diffusion Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Many reward-based fine-tuning methods for diffusion and flow models reduce to a single score-matching objective against a value-guided target.","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Jeongjae Lee, Jeongsol Kim, Jinho Chang, Jong Chul Ye","submitted_at":"2026-04-19T12:47:52Z","abstract_excerpt":"Reward-based fine-tuning steers a pretrained diffusion or flow-based generative model toward higher-reward samples while remaining close to the pretrained model. Although existing methods are derived from different perspectives, we show that many can be written under a common framework, which we call reward score matching (RSM). Under this view, alignment becomes score matching against a value-guided target, and the main differences across methods reduce to the construction of the value-guidance estimator and the effective optimization strength across timesteps. This unification clarifies the "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Although existing methods are derived from different perspectives, we show that many can be written under a common framework, which we call reward score matching (RSM).","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the primary distinctions among existing methods reduce to the construction of the value-guidance estimator and the effective optimization strength across timesteps, without material loss of generality or overlooked auxiliary mechanisms.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Reward Score Matching unifies reward-based fine-tuning for flow and diffusion models by recasting alignment as score matching to a value-guided target.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Many reward-based fine-tuning methods for diffusion and flow models reduce to a single score-matching objective against a value-guided target.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"4a6e37e38fb316c83f3835af855adce2bbccf2923c582855d1481dae5a07005b"},"source":{"id":"2604.17415","kind":"arxiv","version":3},"verdict":{"id":"be8edd1a-49b9-46d3-ba94-a222182b8113","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T05:30:29.711746Z","strongest_claim":"Although existing methods are derived from different perspectives, we show that many can be written under a common framework, which we call reward score matching (RSM).","one_line_summary":"Reward Score Matching unifies reward-based fine-tuning for flow and diffusion models by recasting alignment as score matching to a value-guided target.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the primary distinctions among existing methods reduce to the construction of the value-guidance estimator and the effective optimization strength across timesteps, without material loss of generality or overlooked auxiliary mechanisms.","pith_extraction_headline":"Many reward-based fine-tuning methods for diffusion and flow models reduce to a single score-matching objective against a value-guided target."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.17415/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":"be8edd1a-49b9-46d3-ba94-a222182b8113"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-02T01:03:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fQDeYYTfel5K2xuIyk9OVloTU62SNhl7V+nWlCpg1lZpSF1A1hDGAuLhmk30EIFWXn+SLLY/gmVhPHQjTsrpAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-06T12:55:07.164459Z"},"content_sha256":"5697b95e8df7243c1c71adecbafb634e3b8e09e0403bbf0b33d66f685b490d08","schema_version":"1.0","event_id":"sha256:5697b95e8df7243c1c71adecbafb634e3b8e09e0403bbf0b33d66f685b490d08"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/IOVTADSAIQNRQIZOM5ICML5RTE/bundle.json","state_url":"https://pith.science/pith/IOVTADSAIQNRQIZOM5ICML5RTE/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/IOVTADSAIQNRQIZOM5ICML5RTE/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":"2026-06-06T12:55:07Z","links":{"resolver":"https://pith.science/pith/IOVTADSAIQNRQIZOM5ICML5RTE","bundle":"https://pith.science/pith/IOVTADSAIQNRQIZOM5ICML5RTE/bundle.json","state":"https://pith.science/pith/IOVTADSAIQNRQIZOM5ICML5RTE/state.json","well_known_bundle":"https://pith.science/.well-known/pith/IOVTADSAIQNRQIZOM5ICML5RTE/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:IOVTADSAIQNRQIZOM5ICML5RTE","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":"965112703ecf90a21ea7fa74bb4649f763107130cbece4ac12a9ddcc00b3251d","cross_cats_sorted":["cs.AI","cs.CV"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-04-19T12:47:52Z","title_canon_sha256":"a9b9c9999d2b319856cd98f141b6c5316e3159c1ff90fcf1fc01e0069613e15f"},"schema_version":"1.0","source":{"id":"2604.17415","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.17415","created_at":"2026-06-02T01:03:47Z"},{"alias_kind":"arxiv_version","alias_value":"2604.17415v3","created_at":"2026-06-02T01:03:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.17415","created_at":"2026-06-02T01:03:47Z"},{"alias_kind":"pith_short_12","alias_value":"IOVTADSAIQNR","created_at":"2026-06-02T01:03:47Z"},{"alias_kind":"pith_short_16","alias_value":"IOVTADSAIQNRQIZO","created_at":"2026-06-02T01:03:47Z"},{"alias_kind":"pith_short_8","alias_value":"IOVTADSA","created_at":"2026-06-02T01:03:47Z"}],"graph_snapshots":[{"event_id":"sha256:5697b95e8df7243c1c71adecbafb634e3b8e09e0403bbf0b33d66f685b490d08","target":"graph","created_at":"2026-06-02T01:03:47Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Although existing methods are derived from different perspectives, we show that many can be written under a common framework, which we call reward score matching (RSM)."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the primary distinctions among existing methods reduce to the construction of the value-guidance estimator and the effective optimization strength across timesteps, without material loss of generality or overlooked auxiliary mechanisms."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Reward Score Matching unifies reward-based fine-tuning for flow and diffusion models by recasting alignment as score matching to a value-guided target."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Many reward-based fine-tuning methods for diffusion and flow models reduce to a single score-matching objective against a value-guided target."}],"snapshot_sha256":"4a6e37e38fb316c83f3835af855adce2bbccf2923c582855d1481dae5a07005b"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2604.17415/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reward-based fine-tuning steers a pretrained diffusion or flow-based generative model toward higher-reward samples while remaining close to the pretrained model. Although existing methods are derived from different perspectives, we show that many can be written under a common framework, which we call reward score matching (RSM). Under this view, alignment becomes score matching against a value-guided target, and the main differences across methods reduce to the construction of the value-guidance estimator and the effective optimization strength across timesteps. This unification clarifies the ","authors_text":"Jeongjae Lee, Jeongsol Kim, Jinho Chang, Jong Chul Ye","cross_cats":["cs.AI","cs.CV"],"headline":"Many reward-based fine-tuning methods for diffusion and flow models reduce to a single score-matching objective against a value-guided target.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-04-19T12:47:52Z","title":"Reward Score Matching: Unifying Reward-based Fine-tuning for Flow and Diffusion Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.17415","kind":"arxiv","version":3},"verdict":{"created_at":"2026-05-10T05:30:29.711746Z","id":"be8edd1a-49b9-46d3-ba94-a222182b8113","model_set":{"reader":"grok-4.3"},"one_line_summary":"Reward Score Matching unifies reward-based fine-tuning for flow and diffusion models by recasting alignment as score matching to a value-guided target.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Many reward-based fine-tuning methods for diffusion and flow models reduce to a single score-matching objective against a value-guided target.","strongest_claim":"Although existing methods are derived from different perspectives, we show that many can be written under a common framework, which we call reward score matching (RSM).","weakest_assumption":"That the primary distinctions among existing methods reduce to the construction of the value-guidance estimator and the effective optimization strength across timesteps, without material loss of generality or overlooked auxiliary mechanisms."}},"verdict_id":"be8edd1a-49b9-46d3-ba94-a222182b8113"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:008857f3884c6345f9ea4cb6142d6061174e62b4d0a095f89e429627ac8101b1","target":"record","created_at":"2026-06-02T01:03:47Z","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":"965112703ecf90a21ea7fa74bb4649f763107130cbece4ac12a9ddcc00b3251d","cross_cats_sorted":["cs.AI","cs.CV"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-04-19T12:47:52Z","title_canon_sha256":"a9b9c9999d2b319856cd98f141b6c5316e3159c1ff90fcf1fc01e0069613e15f"},"schema_version":"1.0","source":{"id":"2604.17415","kind":"arxiv","version":3}},"canonical_sha256":"43ab300e40441b18232e6750262fb19903d11d9b06c5390de10b5769fee83e8f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"43ab300e40441b18232e6750262fb19903d11d9b06c5390de10b5769fee83e8f","first_computed_at":"2026-06-02T01:03:47.239586Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T01:03:47.239586Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"stWrP+g57acQKZrSwwom9zJ9sJUEl99Du3ZocIUXw5N3CFk70zaXftmv206rNfz+7dULLWpTV2bfZgmKU2VGAg==","signature_status":"signed_v1","signed_at":"2026-06-02T01:03:47.240129Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.17415","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:008857f3884c6345f9ea4cb6142d6061174e62b4d0a095f89e429627ac8101b1","sha256:5697b95e8df7243c1c71adecbafb634e3b8e09e0403bbf0b33d66f685b490d08"],"state_sha256":"e18244bf465a58b61c7b71acb6d4615eeb7f9c725b65c969416e393c76e7a187"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"daIrXLxhQOvyNyHYGIs2HT1SyYMTTPbARGkhy/7rD1aHZnKql/diL9yaGkI5OgVXE+AWGyngRjMf0vnVMDfxDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-06T12:55:07.166705Z","bundle_sha256":"ddfac28c8df45e6c61d1279a67f04bc97ec251f101099f9322b9da847ed2c68d"}}