{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:OC6VL47P3IO4BZ3NB2GNTKB5VA","short_pith_number":"pith:OC6VL47P","canonical_record":{"source":{"id":"2607.02291","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-02T15:08:56Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"d7e2382728ea5e215723108db16885db5a1c39ce04926e214e3c143906af238a","abstract_canon_sha256":"b425edd6959e0a13ce03bf44c6c64424f5765c5a5c825f9f6f6fe351d7c2c633"},"schema_version":"1.0"},"canonical_sha256":"70bd55f3efda1dc0e76d0e8cd9a83da80c88db7f484a2a003993155d3b4928ff","source":{"kind":"arxiv","id":"2607.02291","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.02291","created_at":"2026-07-03T01:17:47Z"},{"alias_kind":"arxiv_version","alias_value":"2607.02291v1","created_at":"2026-07-03T01:17:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.02291","created_at":"2026-07-03T01:17:47Z"},{"alias_kind":"pith_short_12","alias_value":"OC6VL47P3IO4","created_at":"2026-07-03T01:17:47Z"},{"alias_kind":"pith_short_16","alias_value":"OC6VL47P3IO4BZ3N","created_at":"2026-07-03T01:17:47Z"},{"alias_kind":"pith_short_8","alias_value":"OC6VL47P","created_at":"2026-07-03T01:17:47Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:OC6VL47P3IO4BZ3NB2GNTKB5VA","target":"record","payload":{"canonical_record":{"source":{"id":"2607.02291","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-02T15:08:56Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"d7e2382728ea5e215723108db16885db5a1c39ce04926e214e3c143906af238a","abstract_canon_sha256":"b425edd6959e0a13ce03bf44c6c64424f5765c5a5c825f9f6f6fe351d7c2c633"},"schema_version":"1.0"},"canonical_sha256":"70bd55f3efda1dc0e76d0e8cd9a83da80c88db7f484a2a003993155d3b4928ff","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-03T01:17:47.671523Z","signature_b64":"ZU3qZxtCOxI9XbY9xftmB9m8JBwzQbC9w2o8scCbg0cnPdhXXiQOM1qUT9L7z+U6yP7htE0gSR7AmxWBXhNcAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"70bd55f3efda1dc0e76d0e8cd9a83da80c88db7f484a2a003993155d3b4928ff","last_reissued_at":"2026-07-03T01:17:47.671098Z","signature_status":"signed_v1","first_computed_at":"2026-07-03T01:17:47.671098Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.02291","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":"2026-07-03T01:17:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0ilYmYLuNYTTPM4/wf3oJgjzgIGvynKCtxY4wKJpsmYeCLEx0YZUq2z/MqlP7G3fnJN4Dg/6NgJNGF10EZYKCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-04T06:33:29.251027Z"},"content_sha256":"064dc03aec2f798d1ad859777a67550a00cb0a36b36cb48e0acb93f9cb7e861d","schema_version":"1.0","event_id":"sha256:064dc03aec2f798d1ad859777a67550a00cb0a36b36cb48e0acb93f9cb7e861d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:OC6VL47P3IO4BZ3NB2GNTKB5VA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Optimizing Visual Generative Models via Distribution-wise Rewards","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Fuli Feng, Han Hu, Leigang Qu, Mengde Xu, Ruihang Li, Shuyang Gu, Wenjie Wang","submitted_at":"2026-07-02T15:08:56Z","abstract_excerpt":"Conventional reinforcement learning strategies for visual generation typically employ sample-wise reward functions, yet this practice frequently results in reward hacking that degrades image diversity and introduces visual anomalies. To address these limitations, we present a novel framework that finetunes generative models using distribution-wise rewards, ensuring better alignment with real-world data distributions. Unlike rewards that evaluate samples individually, distribution-wise reward accounts for the data distribution of the samples, mitigating the mode collapse problem that occurs whe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.02291","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/2607.02291/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":"2026-07-03T01:17:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"r5k1l+J1viAhbxdD4i54PSZh3dG/zsNDpTGCZRv0RnKoRgXlVGrt+60aS09TWI6pXu/KBIkAWFtgu7OKKfvgBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-04T06:33:29.251408Z"},"content_sha256":"72185087e6fe4928bb4ee282fe5731d8e083866d0d0df4b073cd4219b74a4f80","schema_version":"1.0","event_id":"sha256:72185087e6fe4928bb4ee282fe5731d8e083866d0d0df4b073cd4219b74a4f80"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/OC6VL47P3IO4BZ3NB2GNTKB5VA/bundle.json","state_url":"https://pith.science/pith/OC6VL47P3IO4BZ3NB2GNTKB5VA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/OC6VL47P3IO4BZ3NB2GNTKB5VA/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":"2026-07-04T06:33:29Z","links":{"resolver":"https://pith.science/pith/OC6VL47P3IO4BZ3NB2GNTKB5VA","bundle":"https://pith.science/pith/OC6VL47P3IO4BZ3NB2GNTKB5VA/bundle.json","state":"https://pith.science/pith/OC6VL47P3IO4BZ3NB2GNTKB5VA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/OC6VL47P3IO4BZ3NB2GNTKB5VA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:OC6VL47P3IO4BZ3NB2GNTKB5VA","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":"b425edd6959e0a13ce03bf44c6c64424f5765c5a5c825f9f6f6fe351d7c2c633","cross_cats_sorted":["cs.CV"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-02T15:08:56Z","title_canon_sha256":"d7e2382728ea5e215723108db16885db5a1c39ce04926e214e3c143906af238a"},"schema_version":"1.0","source":{"id":"2607.02291","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.02291","created_at":"2026-07-03T01:17:47Z"},{"alias_kind":"arxiv_version","alias_value":"2607.02291v1","created_at":"2026-07-03T01:17:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.02291","created_at":"2026-07-03T01:17:47Z"},{"alias_kind":"pith_short_12","alias_value":"OC6VL47P3IO4","created_at":"2026-07-03T01:17:47Z"},{"alias_kind":"pith_short_16","alias_value":"OC6VL47P3IO4BZ3N","created_at":"2026-07-03T01:17:47Z"},{"alias_kind":"pith_short_8","alias_value":"OC6VL47P","created_at":"2026-07-03T01:17:47Z"}],"graph_snapshots":[{"event_id":"sha256:72185087e6fe4928bb4ee282fe5731d8e083866d0d0df4b073cd4219b74a4f80","target":"graph","created_at":"2026-07-03T01:17:47Z","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/2607.02291/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Conventional reinforcement learning strategies for visual generation typically employ sample-wise reward functions, yet this practice frequently results in reward hacking that degrades image diversity and introduces visual anomalies. To address these limitations, we present a novel framework that finetunes generative models using distribution-wise rewards, ensuring better alignment with real-world data distributions. Unlike rewards that evaluate samples individually, distribution-wise reward accounts for the data distribution of the samples, mitigating the mode collapse problem that occurs whe","authors_text":"Fuli Feng, Han Hu, Leigang Qu, Mengde Xu, Ruihang Li, Shuyang Gu, Wenjie Wang","cross_cats":["cs.CV"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-02T15:08:56Z","title":"Optimizing Visual Generative Models via Distribution-wise Rewards"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.02291","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":"sha256:064dc03aec2f798d1ad859777a67550a00cb0a36b36cb48e0acb93f9cb7e861d","target":"record","created_at":"2026-07-03T01:17:47Z","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":"b425edd6959e0a13ce03bf44c6c64424f5765c5a5c825f9f6f6fe351d7c2c633","cross_cats_sorted":["cs.CV"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-02T15:08:56Z","title_canon_sha256":"d7e2382728ea5e215723108db16885db5a1c39ce04926e214e3c143906af238a"},"schema_version":"1.0","source":{"id":"2607.02291","kind":"arxiv","version":1}},"canonical_sha256":"70bd55f3efda1dc0e76d0e8cd9a83da80c88db7f484a2a003993155d3b4928ff","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"70bd55f3efda1dc0e76d0e8cd9a83da80c88db7f484a2a003993155d3b4928ff","first_computed_at":"2026-07-03T01:17:47.671098Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-03T01:17:47.671098Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ZU3qZxtCOxI9XbY9xftmB9m8JBwzQbC9w2o8scCbg0cnPdhXXiQOM1qUT9L7z+U6yP7htE0gSR7AmxWBXhNcAQ==","signature_status":"signed_v1","signed_at":"2026-07-03T01:17:47.671523Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.02291","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:064dc03aec2f798d1ad859777a67550a00cb0a36b36cb48e0acb93f9cb7e861d","sha256:72185087e6fe4928bb4ee282fe5731d8e083866d0d0df4b073cd4219b74a4f80"],"state_sha256":"dff7352b0216802dfa65f6a98a94ca6d46c38582d5f2bb80a7112a5f580408a8"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"a9UZ9x0xxsLrG8uRA6V9jx2jTeNvC+si/aO6JI9BnFcWf6UkVhQIuqbmdOBQQkygv09H1Bp6AFECnb5pyrCvCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-04T06:33:29.253328Z","bundle_sha256":"fff6239fcd77334fd29af66d65be72042fcacbfe22111294ace78176279c6c4e"}}