{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:ULEM7OORTBX2GWADLFPRQCHGVF","short_pith_number":"pith:ULEM7OOR","schema_version":"1.0","canonical_sha256":"a2c8cfb9d1986fa35803595f1808e6a94b3839b544dfc485716044fc76e2c17b","source":{"kind":"arxiv","id":"2404.19753","version":1},"attestation_state":"computed","paper":{"title":"DOCCI: Descriptions of Connected and Contrasting Images","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Alexander Ku, Garrett Tanzer, Jaemin Cho, Jason Baldridge, Jordi Pont-Tuset, Roopal Garg, Sunayana Rane, Su Wang, Yasumasa Onoe, Yonatan Bitton, Zachary Berger, Zarana Parekh","submitted_at":"2024-04-30T17:56:24Z","abstract_excerpt":"Vision-language datasets are vital for both text-to-image (T2I) and image-to-text (I2T) research. However, current datasets lack descriptions with fine-grained detail that would allow for richer associations to be learned by models. To fill the gap, we introduce Descriptions of Connected and Contrasting Images (DOCCI), a dataset with long, human-annotated English descriptions for 15k images that were taken, curated and donated by a single researcher intent on capturing key challenges such as spatial relations, counting, text rendering, world knowledge, and more. We instruct human annotators to"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2404.19753","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-04-30T17:56:24Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"468d890ac2ca152a98776ea58a1a5c980bcdd4f5c3b5eccb08e8617fc5f73596","abstract_canon_sha256":"b09330b52e11305676840d7577cb8658d49221b6f6b4602f75e86f274078e309"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:13:56.401117Z","signature_b64":"hCjgXcH/0ZGQrj7Ou2XtouB+4C7qDds/ttaOYCzjGRCin2HKBxF4lQU63QNoN3Zsbnhh5KFx9KmXxpAzEKGkCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a2c8cfb9d1986fa35803595f1808e6a94b3839b544dfc485716044fc76e2c17b","last_reissued_at":"2026-07-05T08:13:56.400652Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:13:56.400652Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DOCCI: Descriptions of Connected and Contrasting Images","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Alexander Ku, Garrett Tanzer, Jaemin Cho, Jason Baldridge, Jordi Pont-Tuset, Roopal Garg, Sunayana Rane, Su Wang, Yasumasa Onoe, Yonatan Bitton, Zachary Berger, Zarana Parekh","submitted_at":"2024-04-30T17:56:24Z","abstract_excerpt":"Vision-language datasets are vital for both text-to-image (T2I) and image-to-text (I2T) research. However, current datasets lack descriptions with fine-grained detail that would allow for richer associations to be learned by models. To fill the gap, we introduce Descriptions of Connected and Contrasting Images (DOCCI), a dataset with long, human-annotated English descriptions for 15k images that were taken, curated and donated by a single researcher intent on capturing key challenges such as spatial relations, counting, text rendering, world knowledge, and more. We instruct human annotators to"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.19753","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/2404.19753/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2404.19753","created_at":"2026-07-05T08:13:56.400714+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.19753v1","created_at":"2026-07-05T08:13:56.400714+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.19753","created_at":"2026-07-05T08:13:56.400714+00:00"},{"alias_kind":"pith_short_12","alias_value":"ULEM7OORTBX2","created_at":"2026-07-05T08:13:56.400714+00:00"},{"alias_kind":"pith_short_16","alias_value":"ULEM7OORTBX2GWAD","created_at":"2026-07-05T08:13:56.400714+00:00"},{"alias_kind":"pith_short_8","alias_value":"ULEM7OOR","created_at":"2026-07-05T08:13:56.400714+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.27295","citing_title":"Gemini Embedding 2: A Native Multimodal Embedding Model from Gemini","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2506.18871","citing_title":"OmniGen2: Towards Instruction-Aligned Multimodal Generation","ref_index":50,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12357","citing_title":"ReflectCAP: Detailed Image Captioning with Reflective Memory","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2604.05623","citing_title":"DetailVerifyBench: A Benchmark for Dense Hallucination Localization in Long Image Captions","ref_index":33,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ULEM7OORTBX2GWADLFPRQCHGVF","json":"https://pith.science/pith/ULEM7OORTBX2GWADLFPRQCHGVF.json","graph_json":"https://pith.science/api/pith-number/ULEM7OORTBX2GWADLFPRQCHGVF/graph.json","events_json":"https://pith.science/api/pith-number/ULEM7OORTBX2GWADLFPRQCHGVF/events.json","paper":"https://pith.science/paper/ULEM7OOR"},"agent_actions":{"view_html":"https://pith.science/pith/ULEM7OORTBX2GWADLFPRQCHGVF","download_json":"https://pith.science/pith/ULEM7OORTBX2GWADLFPRQCHGVF.json","view_paper":"https://pith.science/paper/ULEM7OOR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.19753&json=true","fetch_graph":"https://pith.science/api/pith-number/ULEM7OORTBX2GWADLFPRQCHGVF/graph.json","fetch_events":"https://pith.science/api/pith-number/ULEM7OORTBX2GWADLFPRQCHGVF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ULEM7OORTBX2GWADLFPRQCHGVF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ULEM7OORTBX2GWADLFPRQCHGVF/action/storage_attestation","attest_author":"https://pith.science/pith/ULEM7OORTBX2GWADLFPRQCHGVF/action/author_attestation","sign_citation":"https://pith.science/pith/ULEM7OORTBX2GWADLFPRQCHGVF/action/citation_signature","submit_replication":"https://pith.science/pith/ULEM7OORTBX2GWADLFPRQCHGVF/action/replication_record"}},"created_at":"2026-07-05T08:13:56.400714+00:00","updated_at":"2026-07-05T08:13:56.400714+00:00"}