{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:65SVTELMDUSKNL35DEB3GUA6OK","short_pith_number":"pith:65SVTELM","schema_version":"1.0","canonical_sha256":"f76559916c1d24a6af7d1903b3501e7282a622eddaa9a50960df4c60c647e070","source":{"kind":"arxiv","id":"2508.07493","version":2},"attestation_state":"computed","paper":{"title":"VisR-Bench: An Empirical Study on Visual Retrieval-Augmented Generation for Multilingual Long Document Understanding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Changyou Chen, Chenguang Wang, Jian Chen, Jihyung Kil, Ming Li, Ruiyi Zhang, Ryan Rossi, Tianyi Zhou, Tong Yu","submitted_at":"2025-08-10T21:44:43Z","abstract_excerpt":"Most organizational data in this world are stored as documents, and visual retrieval plays a crucial role in unlocking the collective intelligence from all these documents. However, existing benchmarks focus on English-only document retrieval or only consider multilingual question-answering on a single-page image. To bridge this gap, we introduce VisR-Bench, a multilingual benchmark designed for question-driven multimodal retrieval in long documents. Our benchmark comprises over 35K high-quality QA pairs across 1.2K documents, enabling fine-grained evaluation of multimodal retrieval. VisR-Benc"},"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":"2508.07493","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-08-10T21:44:43Z","cross_cats_sorted":[],"title_canon_sha256":"d6a68b8e8ddd8ca5419f53357e9fae0e901abd8362a2381c69a41f2500468053","abstract_canon_sha256":"6f08f904e418dde2a282a90e26ce5be22fcba298dd3181f70f1f966f7aa2c5d0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:58:34.370171Z","signature_b64":"K0xxYIyAjv7EZiCUfZaYvTT5+OhfntItu88c/PUilcZUQLihMVTu1xK7Sr6Y9I/F24SzCpS+BRA0t78TdSrkAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f76559916c1d24a6af7d1903b3501e7282a622eddaa9a50960df4c60c647e070","last_reissued_at":"2026-07-05T11:58:34.369718Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:58:34.369718Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VisR-Bench: An Empirical Study on Visual Retrieval-Augmented Generation for Multilingual Long Document Understanding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Changyou Chen, Chenguang Wang, Jian Chen, Jihyung Kil, Ming Li, Ruiyi Zhang, Ryan Rossi, Tianyi Zhou, Tong Yu","submitted_at":"2025-08-10T21:44:43Z","abstract_excerpt":"Most organizational data in this world are stored as documents, and visual retrieval plays a crucial role in unlocking the collective intelligence from all these documents. However, existing benchmarks focus on English-only document retrieval or only consider multilingual question-answering on a single-page image. To bridge this gap, we introduce VisR-Bench, a multilingual benchmark designed for question-driven multimodal retrieval in long documents. Our benchmark comprises over 35K high-quality QA pairs across 1.2K documents, enabling fine-grained evaluation of multimodal retrieval. VisR-Benc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.07493","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/2508.07493/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":"2508.07493","created_at":"2026-07-05T11:58:34.369774+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.07493v2","created_at":"2026-07-05T11:58:34.369774+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.07493","created_at":"2026-07-05T11:58:34.369774+00:00"},{"alias_kind":"pith_short_12","alias_value":"65SVTELMDUSK","created_at":"2026-07-05T11:58:34.369774+00:00"},{"alias_kind":"pith_short_16","alias_value":"65SVTELMDUSKNL35","created_at":"2026-07-05T11:58:34.369774+00:00"},{"alias_kind":"pith_short_8","alias_value":"65SVTELM","created_at":"2026-07-05T11:58:34.369774+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.05927","citing_title":"CMDR: Contextual Multimodal Document Retrieval","ref_index":6,"is_internal_anchor":true},{"citing_arxiv_id":"2606.26122","citing_title":"DocArena: Turning Raw Documents into Controllable Training Environments for Document Search Agents","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06460","citing_title":"MINER: Mining Multimodal Internal Representation for Efficient Retrieval","ref_index":22,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/65SVTELMDUSKNL35DEB3GUA6OK","json":"https://pith.science/pith/65SVTELMDUSKNL35DEB3GUA6OK.json","graph_json":"https://pith.science/api/pith-number/65SVTELMDUSKNL35DEB3GUA6OK/graph.json","events_json":"https://pith.science/api/pith-number/65SVTELMDUSKNL35DEB3GUA6OK/events.json","paper":"https://pith.science/paper/65SVTELM"},"agent_actions":{"view_html":"https://pith.science/pith/65SVTELMDUSKNL35DEB3GUA6OK","download_json":"https://pith.science/pith/65SVTELMDUSKNL35DEB3GUA6OK.json","view_paper":"https://pith.science/paper/65SVTELM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.07493&json=true","fetch_graph":"https://pith.science/api/pith-number/65SVTELMDUSKNL35DEB3GUA6OK/graph.json","fetch_events":"https://pith.science/api/pith-number/65SVTELMDUSKNL35DEB3GUA6OK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/65SVTELMDUSKNL35DEB3GUA6OK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/65SVTELMDUSKNL35DEB3GUA6OK/action/storage_attestation","attest_author":"https://pith.science/pith/65SVTELMDUSKNL35DEB3GUA6OK/action/author_attestation","sign_citation":"https://pith.science/pith/65SVTELMDUSKNL35DEB3GUA6OK/action/citation_signature","submit_replication":"https://pith.science/pith/65SVTELMDUSKNL35DEB3GUA6OK/action/replication_record"}},"created_at":"2026-07-05T11:58:34.369774+00:00","updated_at":"2026-07-05T11:58:34.369774+00:00"}