{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:CXO43YPFRXPFTJIECON3EY74J6","short_pith_number":"pith:CXO43YPF","canonical_record":{"source":{"id":"2404.16820","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-04-25T17:58:43Z","cross_cats_sorted":[],"title_canon_sha256":"304a9e590fe8978c19ce72f4345469b46e26e6d40d5df7c0af93c20e3d69ffb1","abstract_canon_sha256":"336fcbb896b6618215f1fb8c8dc34c1e0d1324cfc9e3e486bf9bdcbd2f1891c4"},"schema_version":"1.0"},"canonical_sha256":"15ddcde1e58dde59a504139bb263fc4f85930a4ee802e77110aefb4a0185f791","source":{"kind":"arxiv","id":"2404.16820","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.16820","created_at":"2026-07-05T10:32:26Z"},{"alias_kind":"arxiv_version","alias_value":"2404.16820v4","created_at":"2026-07-05T10:32:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.16820","created_at":"2026-07-05T10:32:26Z"},{"alias_kind":"pith_short_12","alias_value":"CXO43YPFRXPF","created_at":"2026-07-05T10:32:26Z"},{"alias_kind":"pith_short_16","alias_value":"CXO43YPFRXPFTJIE","created_at":"2026-07-05T10:32:26Z"},{"alias_kind":"pith_short_8","alias_value":"CXO43YPF","created_at":"2026-07-05T10:32:26Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:CXO43YPFRXPFTJIECON3EY74J6","target":"record","payload":{"canonical_record":{"source":{"id":"2404.16820","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-04-25T17:58:43Z","cross_cats_sorted":[],"title_canon_sha256":"304a9e590fe8978c19ce72f4345469b46e26e6d40d5df7c0af93c20e3d69ffb1","abstract_canon_sha256":"336fcbb896b6618215f1fb8c8dc34c1e0d1324cfc9e3e486bf9bdcbd2f1891c4"},"schema_version":"1.0"},"canonical_sha256":"15ddcde1e58dde59a504139bb263fc4f85930a4ee802e77110aefb4a0185f791","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:32:26.294251Z","signature_b64":"dOm+QgUkq1AGWLjs6FNl+HidOR5rzPl9lsusyWGhzDUU4z9YmBOLl1vXVs4+aKbkQ228utWK1tJSxX/mTesDAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"15ddcde1e58dde59a504139bb263fc4f85930a4ee802e77110aefb4a0185f791","last_reissued_at":"2026-07-05T10:32:26.293190Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:32:26.293190Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2404.16820","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:32:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3qsSkiJG9UKurByhbSzy5j11oufJ0ncZKT+jrFrpk1kMZLHjJR+O0TjiQUEk5lsx2LzEMs7EVaiio1BoXWBVBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T13:40:32.470397Z"},"content_sha256":"96acc31d5e9196c7cba0404060a02f3ad91e9f950e32d89c8ed9f22a9230ad98","schema_version":"1.0","event_id":"sha256:96acc31d5e9196c7cba0404060a02f3ad91e9f950e32d89c8ed9f22a9230ad98"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:CXO43YPFRXPFTJIECON3EY74J6","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Revisiting Text-to-Image Evaluation with Gecko: On Metrics, Prompts, and Human Ratings","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Aida Nematzadeh, Anant Nawalgaria, Chris Knutsen, Chuhan Zhang, Cyrus Rashtchian, Emanuele Bugliarello, Ira Ktena, Isabela Albuquerque, Ivana Kaji\\'c, Jordi Pont-Tuset, Olivia Wiles, Pinelopi Papalampidi, Su Wang, Yasumasa Onoe","submitted_at":"2024-04-25T17:58:43Z","abstract_excerpt":"While text-to-image (T2I) generative models have become ubiquitous, they do not necessarily generate images that align with a given prompt. While previous work has evaluated T2I alignment by proposing metrics, benchmarks, and templates for collecting human judgements, the quality of these components is not systematically measured. Human-rated prompt sets are generally small and the reliability of the ratings -- and thereby the prompt set used to compare models -- is not evaluated. We address this gap by performing an extensive study evaluating auto-eval metrics and human templates. We provide "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.16820","kind":"arxiv","version":4},"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/2404.16820/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":"2026-07-05T10:32:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9dY8sB7R5Re5ckuW8UTeewYlVJ8AYuzSWA1kwSY4LhXsAu3G2JM6tiYgcuB321BAnmmu4zzpgxbFPEoT7sMuCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T13:40:32.470901Z"},"content_sha256":"97c83905b12b4ca8b74485b0316515f84781665b73517f47e6b99f212d88f235","schema_version":"1.0","event_id":"sha256:97c83905b12b4ca8b74485b0316515f84781665b73517f47e6b99f212d88f235"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/CXO43YPFRXPFTJIECON3EY74J6/bundle.json","state_url":"https://pith.science/pith/CXO43YPFRXPFTJIECON3EY74J6/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/CXO43YPFRXPFTJIECON3EY74J6/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":"2026-08-08T13:40:32Z","links":{"resolver":"https://pith.science/pith/CXO43YPFRXPFTJIECON3EY74J6","bundle":"https://pith.science/pith/CXO43YPFRXPFTJIECON3EY74J6/bundle.json","state":"https://pith.science/pith/CXO43YPFRXPFTJIECON3EY74J6/state.json","well_known_bundle":"https://pith.science/.well-known/pith/CXO43YPFRXPFTJIECON3EY74J6/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:CXO43YPFRXPFTJIECON3EY74J6","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":"336fcbb896b6618215f1fb8c8dc34c1e0d1324cfc9e3e486bf9bdcbd2f1891c4","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-04-25T17:58:43Z","title_canon_sha256":"304a9e590fe8978c19ce72f4345469b46e26e6d40d5df7c0af93c20e3d69ffb1"},"schema_version":"1.0","source":{"id":"2404.16820","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.16820","created_at":"2026-07-05T10:32:26Z"},{"alias_kind":"arxiv_version","alias_value":"2404.16820v4","created_at":"2026-07-05T10:32:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.16820","created_at":"2026-07-05T10:32:26Z"},{"alias_kind":"pith_short_12","alias_value":"CXO43YPFRXPF","created_at":"2026-07-05T10:32:26Z"},{"alias_kind":"pith_short_16","alias_value":"CXO43YPFRXPFTJIE","created_at":"2026-07-05T10:32:26Z"},{"alias_kind":"pith_short_8","alias_value":"CXO43YPF","created_at":"2026-07-05T10:32:26Z"}],"graph_snapshots":[{"event_id":"sha256:97c83905b12b4ca8b74485b0316515f84781665b73517f47e6b99f212d88f235","target":"graph","created_at":"2026-07-05T10:32:26Z","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/2404.16820/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"While text-to-image (T2I) generative models have become ubiquitous, they do not necessarily generate images that align with a given prompt. While previous work has evaluated T2I alignment by proposing metrics, benchmarks, and templates for collecting human judgements, the quality of these components is not systematically measured. Human-rated prompt sets are generally small and the reliability of the ratings -- and thereby the prompt set used to compare models -- is not evaluated. We address this gap by performing an extensive study evaluating auto-eval metrics and human templates. We provide ","authors_text":"Aida Nematzadeh, Anant Nawalgaria, Chris Knutsen, Chuhan Zhang, Cyrus Rashtchian, Emanuele Bugliarello, Ira Ktena, Isabela Albuquerque, Ivana Kaji\\'c, Jordi Pont-Tuset, Olivia Wiles, Pinelopi Papalampidi, Su Wang, Yasumasa Onoe","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-04-25T17:58:43Z","title":"Revisiting Text-to-Image Evaluation with Gecko: On Metrics, Prompts, and Human Ratings"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.16820","kind":"arxiv","version":4},"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":"sha256:96acc31d5e9196c7cba0404060a02f3ad91e9f950e32d89c8ed9f22a9230ad98","target":"record","created_at":"2026-07-05T10:32:26Z","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":"336fcbb896b6618215f1fb8c8dc34c1e0d1324cfc9e3e486bf9bdcbd2f1891c4","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-04-25T17:58:43Z","title_canon_sha256":"304a9e590fe8978c19ce72f4345469b46e26e6d40d5df7c0af93c20e3d69ffb1"},"schema_version":"1.0","source":{"id":"2404.16820","kind":"arxiv","version":4}},"canonical_sha256":"15ddcde1e58dde59a504139bb263fc4f85930a4ee802e77110aefb4a0185f791","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"15ddcde1e58dde59a504139bb263fc4f85930a4ee802e77110aefb4a0185f791","first_computed_at":"2026-07-05T10:32:26.293190Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:32:26.293190Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"dOm+QgUkq1AGWLjs6FNl+HidOR5rzPl9lsusyWGhzDUU4z9YmBOLl1vXVs4+aKbkQ228utWK1tJSxX/mTesDAA==","signature_status":"signed_v1","signed_at":"2026-07-05T10:32:26.294251Z","signed_message":"canonical_sha256_bytes"},"source_id":"2404.16820","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:96acc31d5e9196c7cba0404060a02f3ad91e9f950e32d89c8ed9f22a9230ad98","sha256:97c83905b12b4ca8b74485b0316515f84781665b73517f47e6b99f212d88f235"],"state_sha256":"b110f7c6c67c5e6119ab85574253cae99aab1ea73a4ab5f259bb9b15e86a9b25"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TnJsmS6y9xsxs8A34acdMs5uldo5w83J0rejoy5R+gNJGVLroC7tX2TJIJsmXqYM2PDibQpyPT1QomsY/meiDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T13:40:32.475720Z","bundle_sha256":"05d9bd8dea4aa1958d0e14e0c7ed16c460524ebc8cfd4396962b0f07614ebbad"}}