{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:BTNBKLKGBC5WQ6OASRDFUAGLBK","short_pith_number":"pith:BTNBKLKG","schema_version":"1.0","canonical_sha256":"0cda152d4608bb6879c094465a00cb0a9f121f24ce3d9acc639745158c6dcaf9","source":{"kind":"arxiv","id":"2405.02793","version":2},"attestation_state":"computed","paper":{"title":"ImageInWords: Unlocking Hyper-Detailed Image Descriptions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Andrea Burns, Andrew Bunner, Burcu Karagol Ayan, Ceslee Montgomery, Jason Baldridge, Radu Soricut, Ranjay Krishna, Roopal Garg, Yasumasa Onoe, Yonatan Bitton","submitted_at":"2024-05-05T02:15:11Z","abstract_excerpt":"Despite the longstanding adage \"an image is worth a thousand words,\" generating accurate hyper-detailed image descriptions remains unsolved. Trained on short web-scraped image text, vision-language models often generate incomplete descriptions with visual inconsistencies. We address this via a novel data-centric approach with ImageInWords (IIW), a carefully designed human-in-the-loop framework for curating hyper-detailed image descriptions. Human evaluations on IIW data show major gains compared to recent datasets (+66%) and GPT4V (+48%) across comprehensiveness, specificity, hallucinations, a"},"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":"2405.02793","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-05-05T02:15:11Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"1d95c508ad1d57f302ca92917a8631e5c9433473da7abf7d5be08594743d8715","abstract_canon_sha256":"7100333a9aa7c0262c523eeb84e4c5b5f2f92e8f3ea512b001771d9f30d348aa"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:27:21.725282Z","signature_b64":"qiotNaYBv+mv079PN1cKeZj29ZfvEncY9h1QvyIJClZhLKzWJaJ96fEpHWJmcCcFAVuB+BbHWe7v1fxKsemMCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0cda152d4608bb6879c094465a00cb0a9f121f24ce3d9acc639745158c6dcaf9","last_reissued_at":"2026-07-05T09:27:21.724751Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:27:21.724751Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ImageInWords: Unlocking Hyper-Detailed Image Descriptions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Andrea Burns, Andrew Bunner, Burcu Karagol Ayan, Ceslee Montgomery, Jason Baldridge, Radu Soricut, Ranjay Krishna, Roopal Garg, Yasumasa Onoe, Yonatan Bitton","submitted_at":"2024-05-05T02:15:11Z","abstract_excerpt":"Despite the longstanding adage \"an image is worth a thousand words,\" generating accurate hyper-detailed image descriptions remains unsolved. Trained on short web-scraped image text, vision-language models often generate incomplete descriptions with visual inconsistencies. We address this via a novel data-centric approach with ImageInWords (IIW), a carefully designed human-in-the-loop framework for curating hyper-detailed image descriptions. Human evaluations on IIW data show major gains compared to recent datasets (+66%) and GPT4V (+48%) across comprehensiveness, specificity, hallucinations, a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.02793","kind":"arxiv","version":2},"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/2405.02793/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":"2405.02793","created_at":"2026-07-05T09:27:21.724819+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.02793v2","created_at":"2026-07-05T09:27:21.724819+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.02793","created_at":"2026-07-05T09:27:21.724819+00:00"},{"alias_kind":"pith_short_12","alias_value":"BTNBKLKGBC5W","created_at":"2026-07-05T09:27:21.724819+00:00"},{"alias_kind":"pith_short_16","alias_value":"BTNBKLKGBC5WQ6OA","created_at":"2026-07-05T09:27:21.724819+00:00"},{"alias_kind":"pith_short_8","alias_value":"BTNBKLKG","created_at":"2026-07-05T09:27:21.724819+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2509.19207","citing_title":"Long Story Short: Disentangling Compositionality and Long-Caption Understanding in Contrastive VLMs","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BTNBKLKGBC5WQ6OASRDFUAGLBK","json":"https://pith.science/pith/BTNBKLKGBC5WQ6OASRDFUAGLBK.json","graph_json":"https://pith.science/api/pith-number/BTNBKLKGBC5WQ6OASRDFUAGLBK/graph.json","events_json":"https://pith.science/api/pith-number/BTNBKLKGBC5WQ6OASRDFUAGLBK/events.json","paper":"https://pith.science/paper/BTNBKLKG"},"agent_actions":{"view_html":"https://pith.science/pith/BTNBKLKGBC5WQ6OASRDFUAGLBK","download_json":"https://pith.science/pith/BTNBKLKGBC5WQ6OASRDFUAGLBK.json","view_paper":"https://pith.science/paper/BTNBKLKG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.02793&json=true","fetch_graph":"https://pith.science/api/pith-number/BTNBKLKGBC5WQ6OASRDFUAGLBK/graph.json","fetch_events":"https://pith.science/api/pith-number/BTNBKLKGBC5WQ6OASRDFUAGLBK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BTNBKLKGBC5WQ6OASRDFUAGLBK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BTNBKLKGBC5WQ6OASRDFUAGLBK/action/storage_attestation","attest_author":"https://pith.science/pith/BTNBKLKGBC5WQ6OASRDFUAGLBK/action/author_attestation","sign_citation":"https://pith.science/pith/BTNBKLKGBC5WQ6OASRDFUAGLBK/action/citation_signature","submit_replication":"https://pith.science/pith/BTNBKLKGBC5WQ6OASRDFUAGLBK/action/replication_record"}},"created_at":"2026-07-05T09:27:21.724819+00:00","updated_at":"2026-07-05T09:27:21.724819+00:00"}