{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:YFUAIQLKJP5SB7VT5V2DHF3FCR","short_pith_number":"pith:YFUAIQLK","schema_version":"1.0","canonical_sha256":"c16804416a4bfb20feb3ed743397651451833e614fdbddb2adfdf8753e98fa04","source":{"kind":"arxiv","id":"2505.24173","version":1},"attestation_state":"computed","paper":{"title":"DrVD-Bench: Do Vision-Language Models Reason Like Human Doctors in Medical Image Diagnosis?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chuxi Xiao, Haiyang Bian, Lei Wei, Tianhong Zhou, Xuegong Zhang, Yingtao Zhu, Yin Xu","submitted_at":"2025-05-30T03:33:25Z","abstract_excerpt":"Vision-language models (VLMs) exhibit strong zero-shot generalization on natural images and show early promise in interpretable medical image analysis. However, existing benchmarks do not systematically evaluate whether these models truly reason like human clinicians or merely imitate superficial patterns. To address this gap, we propose DrVD-Bench, the first multimodal benchmark for clinical visual reasoning. DrVD-Bench consists of three modules: Visual Evidence Comprehension, Reasoning Trajectory Assessment, and Report Generation Evaluation, comprising a total of 7,789 image-question pairs. "},"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":"2505.24173","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-30T03:33:25Z","cross_cats_sorted":[],"title_canon_sha256":"4d293d4e496c78131185afdc17e54116ad1701d7196f43d9796982b4beab57d5","abstract_canon_sha256":"c1bcee55dfa46b29e4064d0f1c84e056bad0eef1be3d8879ea266257e01c16ec"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:12:43.371261Z","signature_b64":"3ChBZzFeNjYBdLtZ17iPl74tmTJSstTfoayt6QnOOBF301NdaV/lLw+uiqasirjlnKM+jc6VDpUNIABAjec6CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c16804416a4bfb20feb3ed743397651451833e614fdbddb2adfdf8753e98fa04","last_reissued_at":"2026-07-05T11:12:43.370771Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:12:43.370771Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DrVD-Bench: Do Vision-Language Models Reason Like Human Doctors in Medical Image Diagnosis?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chuxi Xiao, Haiyang Bian, Lei Wei, Tianhong Zhou, Xuegong Zhang, Yingtao Zhu, Yin Xu","submitted_at":"2025-05-30T03:33:25Z","abstract_excerpt":"Vision-language models (VLMs) exhibit strong zero-shot generalization on natural images and show early promise in interpretable medical image analysis. However, existing benchmarks do not systematically evaluate whether these models truly reason like human clinicians or merely imitate superficial patterns. To address this gap, we propose DrVD-Bench, the first multimodal benchmark for clinical visual reasoning. DrVD-Bench consists of three modules: Visual Evidence Comprehension, Reasoning Trajectory Assessment, and Report Generation Evaluation, comprising a total of 7,789 image-question pairs. "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.24173","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/2505.24173/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":"2505.24173","created_at":"2026-07-05T11:12:43.370828+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.24173v1","created_at":"2026-07-05T11:12:43.370828+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.24173","created_at":"2026-07-05T11:12:43.370828+00:00"},{"alias_kind":"pith_short_12","alias_value":"YFUAIQLKJP5S","created_at":"2026-07-05T11:12:43.370828+00:00"},{"alias_kind":"pith_short_16","alias_value":"YFUAIQLKJP5SB7VT","created_at":"2026-07-05T11:12:43.370828+00:00"},{"alias_kind":"pith_short_8","alias_value":"YFUAIQLK","created_at":"2026-07-05T11:12:43.370828+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.24492","citing_title":"Med-R2: An Adversarial Benchmark for Evidence-Grounded Reasoning in Medical VLMs","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15777","citing_title":"SaaS-Bench: Can Computer-Use Agents Leverage Real-World SaaS to Solve Professional Workflows?","ref_index":92,"is_internal_anchor":false},{"citing_arxiv_id":"2603.24649","citing_title":"MedOpenClaw and MedFlowBench: Auditing Medical Agents in Full-Study Workflows","ref_index":49,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YFUAIQLKJP5SB7VT5V2DHF3FCR","json":"https://pith.science/pith/YFUAIQLKJP5SB7VT5V2DHF3FCR.json","graph_json":"https://pith.science/api/pith-number/YFUAIQLKJP5SB7VT5V2DHF3FCR/graph.json","events_json":"https://pith.science/api/pith-number/YFUAIQLKJP5SB7VT5V2DHF3FCR/events.json","paper":"https://pith.science/paper/YFUAIQLK"},"agent_actions":{"view_html":"https://pith.science/pith/YFUAIQLKJP5SB7VT5V2DHF3FCR","download_json":"https://pith.science/pith/YFUAIQLKJP5SB7VT5V2DHF3FCR.json","view_paper":"https://pith.science/paper/YFUAIQLK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.24173&json=true","fetch_graph":"https://pith.science/api/pith-number/YFUAIQLKJP5SB7VT5V2DHF3FCR/graph.json","fetch_events":"https://pith.science/api/pith-number/YFUAIQLKJP5SB7VT5V2DHF3FCR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YFUAIQLKJP5SB7VT5V2DHF3FCR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YFUAIQLKJP5SB7VT5V2DHF3FCR/action/storage_attestation","attest_author":"https://pith.science/pith/YFUAIQLKJP5SB7VT5V2DHF3FCR/action/author_attestation","sign_citation":"https://pith.science/pith/YFUAIQLKJP5SB7VT5V2DHF3FCR/action/citation_signature","submit_replication":"https://pith.science/pith/YFUAIQLKJP5SB7VT5V2DHF3FCR/action/replication_record"}},"created_at":"2026-07-05T11:12:43.370828+00:00","updated_at":"2026-07-05T11:12:43.370828+00:00"}