{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:BEBOFXDUDPCNHR225V643FJUS7","short_pith_number":"pith:BEBOFXDU","canonical_record":{"source":{"id":"2502.06061","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-09T22:45:15Z","cross_cats_sorted":["cs.AI","cs.CV","stat.ML"],"title_canon_sha256":"f16f7594e57fed67175826afbd295f3ff4914a984f065c103c15f314eafd7b49","abstract_canon_sha256":"391b80b04bb535a7e0c92f169ad9f5e6b9ab662c31e56fac3ad5b9ae47e2ee47"},"schema_version":"1.0"},"canonical_sha256":"0902e2dc741bc4d3c75aed7dcd953497ec64e00086807454a3f2e6ce610f9a5f","source":{"kind":"arxiv","id":"2502.06061","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.06061","created_at":"2026-07-05T10:11:47Z"},{"alias_kind":"arxiv_version","alias_value":"2502.06061v1","created_at":"2026-07-05T10:11:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.06061","created_at":"2026-07-05T10:11:47Z"},{"alias_kind":"pith_short_12","alias_value":"BEBOFXDUDPCN","created_at":"2026-07-05T10:11:47Z"},{"alias_kind":"pith_short_16","alias_value":"BEBOFXDUDPCNHR22","created_at":"2026-07-05T10:11:47Z"},{"alias_kind":"pith_short_8","alias_value":"BEBOFXDU","created_at":"2026-07-05T10:11:47Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:BEBOFXDUDPCNHR225V643FJUS7","target":"record","payload":{"canonical_record":{"source":{"id":"2502.06061","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-09T22:45:15Z","cross_cats_sorted":["cs.AI","cs.CV","stat.ML"],"title_canon_sha256":"f16f7594e57fed67175826afbd295f3ff4914a984f065c103c15f314eafd7b49","abstract_canon_sha256":"391b80b04bb535a7e0c92f169ad9f5e6b9ab662c31e56fac3ad5b9ae47e2ee47"},"schema_version":"1.0"},"canonical_sha256":"0902e2dc741bc4d3c75aed7dcd953497ec64e00086807454a3f2e6ce610f9a5f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:11:47.152808Z","signature_b64":"FhPwMmrTFEXU5EyUFZ8DypHSK65BkQ98mtWa+fXfqujMvjhzXUsLCNuLfD36OX3Bxf3y0MD/gzwoBpzuboWsCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0902e2dc741bc4d3c75aed7dcd953497ec64e00086807454a3f2e6ce610f9a5f","last_reissued_at":"2026-07-05T10:11:47.152337Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:11:47.152337Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2502.06061","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":"2026-07-05T10:11:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"RRCe/euoViQE6tIQXF4jlE1D5wxhU3BQaa4bFjoPHv9GP0vBb3JnJGeQUuqdQADc47J+HiUquecas6FC6NBNCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T00:48:30.829375Z"},"content_sha256":"80660339a01fd77f456f7dcc2107b594502e8684fc615874bd4200183451d9cd","schema_version":"1.0","event_id":"sha256:80660339a01fd77f456f7dcc2107b594502e8684fc615874bd4200183451d9cd"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:BEBOFXDUDPCNHR225V643FJUS7","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Online Reward-Weighted Fine-Tuning of Flow Matching with Wasserstein Regularization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Chaoran Cheng, Chumeng Liang, Ge Liu, Jiajun Fan, Shuaike Shen, Yuxin Chen","submitted_at":"2025-02-09T22:45:15Z","abstract_excerpt":"Recent advancements in reinforcement learning (RL) have achieved great success in fine-tuning diffusion-based generative models. However, fine-tuning continuous flow-based generative models to align with arbitrary user-defined reward functions remains challenging, particularly due to issues such as policy collapse from overoptimization and the prohibitively high computational cost of likelihoods in continuous-time flows. In this paper, we propose an easy-to-use and theoretically sound RL fine-tuning method, which we term Online Reward-Weighted Conditional Flow Matching with Wasserstein-2 Regul"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.06061","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/2502.06061/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":"2026-07-05T10:11:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dozmIFOpXnujsjXCfOub4FHGywe/DThtZSOF4gDfbYQ+9F+QXEGVqAZ/m2Irlcm6cqIU++/IM3wFBz+hchcvAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T00:48:30.829871Z"},"content_sha256":"c006fdcfe9c66492e2bd58ea8edc4b27b60421221dfd2071dd461d5e037d7502","schema_version":"1.0","event_id":"sha256:c006fdcfe9c66492e2bd58ea8edc4b27b60421221dfd2071dd461d5e037d7502"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BEBOFXDUDPCNHR225V643FJUS7/bundle.json","state_url":"https://pith.science/pith/BEBOFXDUDPCNHR225V643FJUS7/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BEBOFXDUDPCNHR225V643FJUS7/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":"2026-08-07T00:48:30Z","links":{"resolver":"https://pith.science/pith/BEBOFXDUDPCNHR225V643FJUS7","bundle":"https://pith.science/pith/BEBOFXDUDPCNHR225V643FJUS7/bundle.json","state":"https://pith.science/pith/BEBOFXDUDPCNHR225V643FJUS7/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BEBOFXDUDPCNHR225V643FJUS7/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:BEBOFXDUDPCNHR225V643FJUS7","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":"391b80b04bb535a7e0c92f169ad9f5e6b9ab662c31e56fac3ad5b9ae47e2ee47","cross_cats_sorted":["cs.AI","cs.CV","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-09T22:45:15Z","title_canon_sha256":"f16f7594e57fed67175826afbd295f3ff4914a984f065c103c15f314eafd7b49"},"schema_version":"1.0","source":{"id":"2502.06061","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.06061","created_at":"2026-07-05T10:11:47Z"},{"alias_kind":"arxiv_version","alias_value":"2502.06061v1","created_at":"2026-07-05T10:11:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.06061","created_at":"2026-07-05T10:11:47Z"},{"alias_kind":"pith_short_12","alias_value":"BEBOFXDUDPCN","created_at":"2026-07-05T10:11:47Z"},{"alias_kind":"pith_short_16","alias_value":"BEBOFXDUDPCNHR22","created_at":"2026-07-05T10:11:47Z"},{"alias_kind":"pith_short_8","alias_value":"BEBOFXDU","created_at":"2026-07-05T10:11:47Z"}],"graph_snapshots":[{"event_id":"sha256:c006fdcfe9c66492e2bd58ea8edc4b27b60421221dfd2071dd461d5e037d7502","target":"graph","created_at":"2026-07-05T10:11:47Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2502.06061/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent advancements in reinforcement learning (RL) have achieved great success in fine-tuning diffusion-based generative models. However, fine-tuning continuous flow-based generative models to align with arbitrary user-defined reward functions remains challenging, particularly due to issues such as policy collapse from overoptimization and the prohibitively high computational cost of likelihoods in continuous-time flows. In this paper, we propose an easy-to-use and theoretically sound RL fine-tuning method, which we term Online Reward-Weighted Conditional Flow Matching with Wasserstein-2 Regul","authors_text":"Chaoran Cheng, Chumeng Liang, Ge Liu, Jiajun Fan, Shuaike Shen, Yuxin Chen","cross_cats":["cs.AI","cs.CV","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-09T22:45:15Z","title":"Online Reward-Weighted Fine-Tuning of Flow Matching with Wasserstein Regularization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.06061","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":"sha256:80660339a01fd77f456f7dcc2107b594502e8684fc615874bd4200183451d9cd","target":"record","created_at":"2026-07-05T10:11:47Z","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":"391b80b04bb535a7e0c92f169ad9f5e6b9ab662c31e56fac3ad5b9ae47e2ee47","cross_cats_sorted":["cs.AI","cs.CV","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-09T22:45:15Z","title_canon_sha256":"f16f7594e57fed67175826afbd295f3ff4914a984f065c103c15f314eafd7b49"},"schema_version":"1.0","source":{"id":"2502.06061","kind":"arxiv","version":1}},"canonical_sha256":"0902e2dc741bc4d3c75aed7dcd953497ec64e00086807454a3f2e6ce610f9a5f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0902e2dc741bc4d3c75aed7dcd953497ec64e00086807454a3f2e6ce610f9a5f","first_computed_at":"2026-07-05T10:11:47.152337Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:11:47.152337Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"FhPwMmrTFEXU5EyUFZ8DypHSK65BkQ98mtWa+fXfqujMvjhzXUsLCNuLfD36OX3Bxf3y0MD/gzwoBpzuboWsCA==","signature_status":"signed_v1","signed_at":"2026-07-05T10:11:47.152808Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.06061","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:80660339a01fd77f456f7dcc2107b594502e8684fc615874bd4200183451d9cd","sha256:c006fdcfe9c66492e2bd58ea8edc4b27b60421221dfd2071dd461d5e037d7502"],"state_sha256":"b1abf36366ac965eb9fca458ad49af253d67ff88cef97cced393dc0ca3ffe0c0"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PAemhbEBB4nZtuWblO4IFOxBckRnSHhf8NcJaEZNRDQU8Yn+MzhcukXI5I0rIRyi1mxw5zIPLnWWZ6Ui2T/3AA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T00:48:30.833420Z","bundle_sha256":"65a69946e3e9dea61189886492c1c17fcb6e728db1c3676b58587a11024f4ed5"}}