{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:MX6NWEHNQZHWHMO36SNMA6THIT","short_pith_number":"pith:MX6NWEHN","canonical_record":{"source":{"id":"2404.01863","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-04-02T11:40:38Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9bf72e7de338665902c0b100bbf6640e2c88e3ad14b27fb9ef344fa3f80a1fd1","abstract_canon_sha256":"d03b4ffb41a8ceab581b1404a3953b86ea15f192c31099c487a46b40fe007e7c"},"schema_version":"1.0"},"canonical_sha256":"65fcdb10ed864f63b1dbf49ac07a6744c3acf118d4ae92c65f983db694285bf4","source":{"kind":"arxiv","id":"2404.01863","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.01863","created_at":"2026-07-05T08:03:32Z"},{"alias_kind":"arxiv_version","alias_value":"2404.01863v1","created_at":"2026-07-05T08:03:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.01863","created_at":"2026-07-05T08:03:32Z"},{"alias_kind":"pith_short_12","alias_value":"MX6NWEHNQZHW","created_at":"2026-07-05T08:03:32Z"},{"alias_kind":"pith_short_16","alias_value":"MX6NWEHNQZHWHMO3","created_at":"2026-07-05T08:03:32Z"},{"alias_kind":"pith_short_8","alias_value":"MX6NWEHN","created_at":"2026-07-05T08:03:32Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:MX6NWEHNQZHWHMO36SNMA6THIT","target":"record","payload":{"canonical_record":{"source":{"id":"2404.01863","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-04-02T11:40:38Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9bf72e7de338665902c0b100bbf6640e2c88e3ad14b27fb9ef344fa3f80a1fd1","abstract_canon_sha256":"d03b4ffb41a8ceab581b1404a3953b86ea15f192c31099c487a46b40fe007e7c"},"schema_version":"1.0"},"canonical_sha256":"65fcdb10ed864f63b1dbf49ac07a6744c3acf118d4ae92c65f983db694285bf4","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:03:32.796154Z","signature_b64":"J3+EZuyq8CzrXy8Bkf5WeZEWr+FGZqdmmLU+GAEeeNevSrfpbUSvU56AzYR2+/mFHJsWi+yZB/I52RL5x8/aCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"65fcdb10ed864f63b1dbf49ac07a6744c3acf118d4ae92c65f983db694285bf4","last_reissued_at":"2026-07-05T08:03:32.795681Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:03:32.795681Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2404.01863","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":"2026-07-05T08:03:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"jDpy6jt/BAFSHM9S1Ew2WttxUJ8fVWuR66Qiu+8s8GH+HYxaFOApjOx6lJBKk5Uc/sRzeScLtS2ENL1qwimQBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T22:48:45.040295Z"},"content_sha256":"7e9acd245b84d0c4f47f629b1dd4eed57d8f21d2afc6be33150b32447f7632b3","schema_version":"1.0","event_id":"sha256:7e9acd245b84d0c4f47f629b1dd4eed57d8f21d2afc6be33150b32447f7632b3"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:MX6NWEHNQZHWHMO36SNMA6THIT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Confidence-aware Reward Optimization for Fine-tuning Text-to-Image Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Jinwoo Shin, Jongheon Jeong, Kimin Lee, Krishnamurthy Dvijotham, Kyuyoung Kim, Minyong An, Mohammad Ghavamzadeh","submitted_at":"2024-04-02T11:40:38Z","abstract_excerpt":"Fine-tuning text-to-image models with reward functions trained on human feedback data has proven effective for aligning model behavior with human intent. However, excessive optimization with such reward models, which serve as mere proxy objectives, can compromise the performance of fine-tuned models, a phenomenon known as reward overoptimization. To investigate this issue in depth, we introduce the Text-Image Alignment Assessment (TIA2) benchmark, which comprises a diverse collection of text prompts, images, and human annotations. Our evaluation of several state-of-the-art reward models on thi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.01863","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/2404.01863/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":"2026-07-05T08:03:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"KANQww3Az9AFHaxR3axpj2BauK7C949ov7sSCF0UXkJhqDsaHmAhNaxxaWKtNtfyDA0ZuorjlNAM3+vaVaDCAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T22:48:45.040852Z"},"content_sha256":"15be90a046fcd513c10e41a869bc101db374a96e9387206cdc58ab693f14f228","schema_version":"1.0","event_id":"sha256:15be90a046fcd513c10e41a869bc101db374a96e9387206cdc58ab693f14f228"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/MX6NWEHNQZHWHMO36SNMA6THIT/bundle.json","state_url":"https://pith.science/pith/MX6NWEHNQZHWHMO36SNMA6THIT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/MX6NWEHNQZHWHMO36SNMA6THIT/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":"2026-08-08T22:48:45Z","links":{"resolver":"https://pith.science/pith/MX6NWEHNQZHWHMO36SNMA6THIT","bundle":"https://pith.science/pith/MX6NWEHNQZHWHMO36SNMA6THIT/bundle.json","state":"https://pith.science/pith/MX6NWEHNQZHWHMO36SNMA6THIT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/MX6NWEHNQZHWHMO36SNMA6THIT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:MX6NWEHNQZHWHMO36SNMA6THIT","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":"d03b4ffb41a8ceab581b1404a3953b86ea15f192c31099c487a46b40fe007e7c","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-04-02T11:40:38Z","title_canon_sha256":"9bf72e7de338665902c0b100bbf6640e2c88e3ad14b27fb9ef344fa3f80a1fd1"},"schema_version":"1.0","source":{"id":"2404.01863","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.01863","created_at":"2026-07-05T08:03:32Z"},{"alias_kind":"arxiv_version","alias_value":"2404.01863v1","created_at":"2026-07-05T08:03:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.01863","created_at":"2026-07-05T08:03:32Z"},{"alias_kind":"pith_short_12","alias_value":"MX6NWEHNQZHW","created_at":"2026-07-05T08:03:32Z"},{"alias_kind":"pith_short_16","alias_value":"MX6NWEHNQZHWHMO3","created_at":"2026-07-05T08:03:32Z"},{"alias_kind":"pith_short_8","alias_value":"MX6NWEHN","created_at":"2026-07-05T08:03:32Z"}],"graph_snapshots":[{"event_id":"sha256:15be90a046fcd513c10e41a869bc101db374a96e9387206cdc58ab693f14f228","target":"graph","created_at":"2026-07-05T08:03:32Z","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/2404.01863/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Fine-tuning text-to-image models with reward functions trained on human feedback data has proven effective for aligning model behavior with human intent. However, excessive optimization with such reward models, which serve as mere proxy objectives, can compromise the performance of fine-tuned models, a phenomenon known as reward overoptimization. To investigate this issue in depth, we introduce the Text-Image Alignment Assessment (TIA2) benchmark, which comprises a diverse collection of text prompts, images, and human annotations. Our evaluation of several state-of-the-art reward models on thi","authors_text":"Jinwoo Shin, Jongheon Jeong, Kimin Lee, Krishnamurthy Dvijotham, Kyuyoung Kim, Minyong An, Mohammad Ghavamzadeh","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-04-02T11:40:38Z","title":"Confidence-aware Reward Optimization for Fine-tuning Text-to-Image Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.01863","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":"sha256:7e9acd245b84d0c4f47f629b1dd4eed57d8f21d2afc6be33150b32447f7632b3","target":"record","created_at":"2026-07-05T08:03:32Z","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":"d03b4ffb41a8ceab581b1404a3953b86ea15f192c31099c487a46b40fe007e7c","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-04-02T11:40:38Z","title_canon_sha256":"9bf72e7de338665902c0b100bbf6640e2c88e3ad14b27fb9ef344fa3f80a1fd1"},"schema_version":"1.0","source":{"id":"2404.01863","kind":"arxiv","version":1}},"canonical_sha256":"65fcdb10ed864f63b1dbf49ac07a6744c3acf118d4ae92c65f983db694285bf4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"65fcdb10ed864f63b1dbf49ac07a6744c3acf118d4ae92c65f983db694285bf4","first_computed_at":"2026-07-05T08:03:32.795681Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:03:32.795681Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"J3+EZuyq8CzrXy8Bkf5WeZEWr+FGZqdmmLU+GAEeeNevSrfpbUSvU56AzYR2+/mFHJsWi+yZB/I52RL5x8/aCg==","signature_status":"signed_v1","signed_at":"2026-07-05T08:03:32.796154Z","signed_message":"canonical_sha256_bytes"},"source_id":"2404.01863","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7e9acd245b84d0c4f47f629b1dd4eed57d8f21d2afc6be33150b32447f7632b3","sha256:15be90a046fcd513c10e41a869bc101db374a96e9387206cdc58ab693f14f228"],"state_sha256":"693122642e477e99f0d388b6a0ff0465eb5898e67a5e500ee7834d7a50241c2a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hAeQj4EcIN/z6vFMGiD1D4NWyol2PZSsxsEJFmL4y9auxjflfiJpF8UZ0uc8v0Xe9B1ntbw5aH+ewW1AF1vgBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T22:48:45.044283Z","bundle_sha256":"f95fcc55fb815f751d3cb5f2fab2faedb8de4639ba2a68bef53c1802c46983cc"}}