{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:3MZKSOADVQZAPTFSGJYSUHF63T","short_pith_number":"pith:3MZKSOAD","schema_version":"1.0","canonical_sha256":"db32a93803ac3207ccb232712a1cbedccf206d8716f6aea330c8c3a3900554a4","source":{"kind":"arxiv","id":"2410.21311","version":1},"attestation_state":"computed","paper":{"title":"MMDocBench: Benchmarking Large Vision-Language Models for Fine-Grained Visual Document Understanding","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Chao Wang, Fengbin Zhu, Fuli Feng, Haohui Wu, Huanbo Luan, Tat Seng Chua, Wenjie Wang, Xiang Yao Ng, Ziyang Liu","submitted_at":"2024-10-25T16:00:55Z","abstract_excerpt":"Large Vision-Language Models (LVLMs) have achieved remarkable performance in many vision-language tasks, yet their capabilities in fine-grained visual understanding remain insufficiently evaluated. Existing benchmarks either contain limited fine-grained evaluation samples that are mixed with other data, or are confined to object-level assessments in natural images. To holistically assess LVLMs' fine-grained visual understanding capabilities, we propose using document images with multi-granularity and multi-modal information to supplement natural images. In this light, we construct MMDocBench, "},"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":"2410.21311","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2024-10-25T16:00:55Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"a4aca1a95a9636a078425906a28ee0f6afe9412aefeb289b87f0089078efcf19","abstract_canon_sha256":"06f44b2a98d057c70ed7bc3ad2638caac716722d68aa426ec2008365a2b7575d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:27:26.583220Z","signature_b64":"Yd50lZMtXNzBcQ2OfvGBIxHHX2iFk82RfPWFGsBulIsOsrvDtF7zSBRfG2WIVSzM/zwtYXfqc22ACWESh9TtDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"db32a93803ac3207ccb232712a1cbedccf206d8716f6aea330c8c3a3900554a4","last_reissued_at":"2026-07-05T09:27:26.582760Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:27:26.582760Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MMDocBench: Benchmarking Large Vision-Language Models for Fine-Grained Visual Document Understanding","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Chao Wang, Fengbin Zhu, Fuli Feng, Haohui Wu, Huanbo Luan, Tat Seng Chua, Wenjie Wang, Xiang Yao Ng, Ziyang Liu","submitted_at":"2024-10-25T16:00:55Z","abstract_excerpt":"Large Vision-Language Models (LVLMs) have achieved remarkable performance in many vision-language tasks, yet their capabilities in fine-grained visual understanding remain insufficiently evaluated. Existing benchmarks either contain limited fine-grained evaluation samples that are mixed with other data, or are confined to object-level assessments in natural images. To holistically assess LVLMs' fine-grained visual understanding capabilities, we propose using document images with multi-granularity and multi-modal information to supplement natural images. In this light, we construct MMDocBench, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.21311","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/2410.21311/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":"2410.21311","created_at":"2026-07-05T09:27:26.582828+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.21311v1","created_at":"2026-07-05T09:27:26.582828+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.21311","created_at":"2026-07-05T09:27:26.582828+00:00"},{"alias_kind":"pith_short_12","alias_value":"3MZKSOADVQZA","created_at":"2026-07-05T09:27:26.582828+00:00"},{"alias_kind":"pith_short_16","alias_value":"3MZKSOADVQZAPTFS","created_at":"2026-07-05T09:27:26.582828+00:00"},{"alias_kind":"pith_short_8","alias_value":"3MZKSOAD","created_at":"2026-07-05T09:27:26.582828+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.01245","citing_title":"Office Comprehension Benchmark","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01936","citing_title":"What to Format and How: A Benchmark and Workflow Approach for Document Formatting","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2606.28344","citing_title":"PIXELRAG: Web Screenshots Beat Text for Retrieval-Augmented Generation","ref_index":46,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3MZKSOADVQZAPTFSGJYSUHF63T","json":"https://pith.science/pith/3MZKSOADVQZAPTFSGJYSUHF63T.json","graph_json":"https://pith.science/api/pith-number/3MZKSOADVQZAPTFSGJYSUHF63T/graph.json","events_json":"https://pith.science/api/pith-number/3MZKSOADVQZAPTFSGJYSUHF63T/events.json","paper":"https://pith.science/paper/3MZKSOAD"},"agent_actions":{"view_html":"https://pith.science/pith/3MZKSOADVQZAPTFSGJYSUHF63T","download_json":"https://pith.science/pith/3MZKSOADVQZAPTFSGJYSUHF63T.json","view_paper":"https://pith.science/paper/3MZKSOAD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.21311&json=true","fetch_graph":"https://pith.science/api/pith-number/3MZKSOADVQZAPTFSGJYSUHF63T/graph.json","fetch_events":"https://pith.science/api/pith-number/3MZKSOADVQZAPTFSGJYSUHF63T/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3MZKSOADVQZAPTFSGJYSUHF63T/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3MZKSOADVQZAPTFSGJYSUHF63T/action/storage_attestation","attest_author":"https://pith.science/pith/3MZKSOADVQZAPTFSGJYSUHF63T/action/author_attestation","sign_citation":"https://pith.science/pith/3MZKSOADVQZAPTFSGJYSUHF63T/action/citation_signature","submit_replication":"https://pith.science/pith/3MZKSOADVQZAPTFSGJYSUHF63T/action/replication_record"}},"created_at":"2026-07-05T09:27:26.582828+00:00","updated_at":"2026-07-05T09:27:26.582828+00:00"}