{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:JE2DQKXLVZTEM3HMN4TWRHTQ5E","short_pith_number":"pith:JE2DQKXL","schema_version":"1.0","canonical_sha256":"4934382aebae66466cec6f27689e70e933c12b2fa0992274f3f1cdadfaae155f","source":{"kind":"arxiv","id":"2602.00593","version":3},"attestation_state":"computed","paper":{"title":"Pix2Fact: When Vision Is Not Enough -- Benchmarking Fine-Grained VQA with Web Verification on High-Resolution Real-World Scenes","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"Current top vision-language models reach only 51.7 percent accuracy on questions that demand both precise visual details from high-resolution scenes and external knowledge verification.","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Bingzhang Wang, Bofei Zhang, Cong Zhang, Qiaofeng Zheng, Yew-Soon Ong, Yifan Jiang, Yifan Yang","submitted_at":"2026-01-31T08:18:34Z","abstract_excerpt":"Despite progress on general tasks, vision-language models (VLMs) still struggle with challenges that demand both fine-grained visual grounding and external knowledge, a synergy overlooked by existing benchmarks that evaluate these abilities in isolation. To fill this void, we introduce Pix2Fact, a visual question-answering benchmark designed to assess expert-level visual perception and knowledge search. Pix2Fact comprises 1,000 high-resolution (4K+) images spanning eight scenarios. Its questions and answers are meticulously crafted by PhD-holding annotators from top global universities across "},"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":"2602.00593","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-01-31T08:18:34Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"f4ac130dec96a28cdd2bfa5a7818e264f2516ede4ee3c2fc286eae329071d30c","abstract_canon_sha256":"9d8ea0c9882e7ffc902416ce4f3798bf7963edf2a08bbb474f7aed1126b7b940"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-21T01:05:15.364638Z","signature_b64":"60qAourmq904RxiIYLnfMyYaNWZJeMl6KkOr8PbRd3IlMkBWWBsoZK7ANnxBgK0R8g0urvjm3BBab2wHECVpCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4934382aebae66466cec6f27689e70e933c12b2fa0992274f3f1cdadfaae155f","last_reissued_at":"2026-05-21T01:05:15.363778Z","signature_status":"signed_v1","first_computed_at":"2026-05-21T01:05:15.363778Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Pix2Fact: When Vision Is Not Enough -- Benchmarking Fine-Grained VQA with Web Verification on High-Resolution Real-World Scenes","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"Current top vision-language models reach only 51.7 percent accuracy on questions that demand both precise visual details from high-resolution scenes and external knowledge verification.","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Bingzhang Wang, Bofei Zhang, Cong Zhang, Qiaofeng Zheng, Yew-Soon Ong, Yifan Jiang, Yifan Yang","submitted_at":"2026-01-31T08:18:34Z","abstract_excerpt":"Despite progress on general tasks, vision-language models (VLMs) still struggle with challenges that demand both fine-grained visual grounding and external knowledge, a synergy overlooked by existing benchmarks that evaluate these abilities in isolation. To fill this void, we introduce Pix2Fact, a visual question-answering benchmark designed to assess expert-level visual perception and knowledge search. Pix2Fact comprises 1,000 high-resolution (4K+) images spanning eight scenarios. Its questions and answers are meticulously crafted by PhD-holding annotators from top global universities across "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"the most advanced model (Gemini-3.1-Pro) achieves only 51.7% average accuracy, even with access to visual ground truth and search tools.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The questions and answers produced by PhD annotators from top universities faithfully represent expert-level challenges of fine-grained visual grounding plus external knowledge without introducing systematic biases or inconsistent difficulty.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Pix2Fact benchmark shows top VLMs achieve only 51.7 percent accuracy on fine-grained visual questions needing both detailed image grounding and web-verified external knowledge.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Current top vision-language models reach only 51.7 percent accuracy on questions that demand both precise visual details from high-resolution scenes and external knowledge verification.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"216b1ca2821a3dd597ac82d45526274301a1869b4601e9ea8532bb2c4b01b06b"},"source":{"id":"2602.00593","kind":"arxiv","version":3},"verdict":{"id":"18d063c2-dd20-497c-8c2a-5621250bc686","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T09:11:44.471030Z","strongest_claim":"the most advanced model (Gemini-3.1-Pro) achieves only 51.7% average accuracy, even with access to visual ground truth and search tools.","one_line_summary":"Pix2Fact benchmark shows top VLMs achieve only 51.7 percent accuracy on fine-grained visual questions needing both detailed image grounding and web-verified external knowledge.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The questions and answers produced by PhD annotators from top universities faithfully represent expert-level challenges of fine-grained visual grounding plus external knowledge without introducing systematic biases or inconsistent difficulty.","pith_extraction_headline":"Current top vision-language models reach only 51.7 percent accuracy on questions that demand both precise visual details from high-resolution scenes and external knowledge verification."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2602.00593/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":"2602.00593","created_at":"2026-05-21T01:05:15.363888+00:00"},{"alias_kind":"arxiv_version","alias_value":"2602.00593v3","created_at":"2026-05-21T01:05:15.363888+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.00593","created_at":"2026-05-21T01:05:15.363888+00:00"},{"alias_kind":"pith_short_12","alias_value":"JE2DQKXLVZTE","created_at":"2026-05-21T01:05:15.363888+00:00"},{"alias_kind":"pith_short_16","alias_value":"JE2DQKXLVZTEM3HM","created_at":"2026-05-21T01:05:15.363888+00:00"},{"alias_kind":"pith_short_8","alias_value":"JE2DQKXL","created_at":"2026-05-21T01:05:15.363888+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2606.05576","citing_title":"UltraVR: A Diagnostic Ultra-Resolution Image-VQA Benchmark for Evidence-Grounded Reasoning","ref_index":14,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JE2DQKXLVZTEM3HMN4TWRHTQ5E","json":"https://pith.science/pith/JE2DQKXLVZTEM3HMN4TWRHTQ5E.json","graph_json":"https://pith.science/api/pith-number/JE2DQKXLVZTEM3HMN4TWRHTQ5E/graph.json","events_json":"https://pith.science/api/pith-number/JE2DQKXLVZTEM3HMN4TWRHTQ5E/events.json","paper":"https://pith.science/paper/JE2DQKXL"},"agent_actions":{"view_html":"https://pith.science/pith/JE2DQKXLVZTEM3HMN4TWRHTQ5E","download_json":"https://pith.science/pith/JE2DQKXLVZTEM3HMN4TWRHTQ5E.json","view_paper":"https://pith.science/paper/JE2DQKXL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2602.00593&json=true","fetch_graph":"https://pith.science/api/pith-number/JE2DQKXLVZTEM3HMN4TWRHTQ5E/graph.json","fetch_events":"https://pith.science/api/pith-number/JE2DQKXLVZTEM3HMN4TWRHTQ5E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JE2DQKXLVZTEM3HMN4TWRHTQ5E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JE2DQKXLVZTEM3HMN4TWRHTQ5E/action/storage_attestation","attest_author":"https://pith.science/pith/JE2DQKXLVZTEM3HMN4TWRHTQ5E/action/author_attestation","sign_citation":"https://pith.science/pith/JE2DQKXLVZTEM3HMN4TWRHTQ5E/action/citation_signature","submit_replication":"https://pith.science/pith/JE2DQKXLVZTEM3HMN4TWRHTQ5E/action/replication_record"}},"created_at":"2026-05-21T01:05:15.363888+00:00","updated_at":"2026-05-21T01:05:15.363888+00:00"}