{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:L6RYV4JJSLKKB4EMRZM6PU3ZHU","short_pith_number":"pith:L6RYV4JJ","schema_version":"1.0","canonical_sha256":"5fa38af12992d4a0f08c8e59e7d3793d317ef0b5b1497dfbbb6363eb5c975218","source":{"kind":"arxiv","id":"2502.18017","version":2},"attestation_state":"computed","paper":{"title":"ViDoRAG: Visual Document Retrieval-Augmented Generation via Dynamic Iterative Reasoning Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.IR"],"primary_cat":"cs.CV","authors_text":"Feng Zhao, Pengjun Xie, Qiuchen Wang, Ruixue Ding, Shihang Wang, Weiqi Wu, Zehui Chen","submitted_at":"2025-02-25T09:26:12Z","abstract_excerpt":"Understanding information from visually rich documents remains a significant challenge for traditional Retrieval-Augmented Generation (RAG) methods. Existing benchmarks predominantly focus on image-based question answering (QA), overlooking the fundamental challenges of efficient retrieval, comprehension, and reasoning within dense visual documents. To bridge this gap, we introduce ViDoSeek, a novel dataset designed to evaluate RAG performance on visually rich documents requiring complex reasoning. Based on it, we identify key limitations in current RAG approaches: (i) purely visual retrieval "},"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":"2502.18017","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-02-25T09:26:12Z","cross_cats_sorted":["cs.AI","cs.CL","cs.IR"],"title_canon_sha256":"64b0d639ee3acd8902c0bb6317117df7a2ccfa23853cc1948149f73f5670f80b","abstract_canon_sha256":"91fa63e3410e6b0788aac2f94d0abc2f7318eb1f1c5831cffc7266c0917dc129"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:14:54.397267Z","signature_b64":"MT/crvvPsV+SiCHS1l/LB7noEz1cQg2foZyuBUtWpEBgFT2+lhQhxf6wZfGVjgWUtLBpStBEOm+o174C99wJCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5fa38af12992d4a0f08c8e59e7d3793d317ef0b5b1497dfbbb6363eb5c975218","last_reissued_at":"2026-07-05T11:14:54.396811Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:14:54.396811Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ViDoRAG: Visual Document Retrieval-Augmented Generation via Dynamic Iterative Reasoning Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.IR"],"primary_cat":"cs.CV","authors_text":"Feng Zhao, Pengjun Xie, Qiuchen Wang, Ruixue Ding, Shihang Wang, Weiqi Wu, Zehui Chen","submitted_at":"2025-02-25T09:26:12Z","abstract_excerpt":"Understanding information from visually rich documents remains a significant challenge for traditional Retrieval-Augmented Generation (RAG) methods. Existing benchmarks predominantly focus on image-based question answering (QA), overlooking the fundamental challenges of efficient retrieval, comprehension, and reasoning within dense visual documents. To bridge this gap, we introduce ViDoSeek, a novel dataset designed to evaluate RAG performance on visually rich documents requiring complex reasoning. Based on it, we identify key limitations in current RAG approaches: (i) purely visual retrieval "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.18017","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/2502.18017/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":"2502.18017","created_at":"2026-07-05T11:14:54.396869+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.18017v2","created_at":"2026-07-05T11:14:54.396869+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.18017","created_at":"2026-07-05T11:14:54.396869+00:00"},{"alias_kind":"pith_short_12","alias_value":"L6RYV4JJSLKK","created_at":"2026-07-05T11:14:54.396869+00:00"},{"alias_kind":"pith_short_16","alias_value":"L6RYV4JJSLKKB4EM","created_at":"2026-07-05T11:14:54.396869+00:00"},{"alias_kind":"pith_short_8","alias_value":"L6RYV4JJ","created_at":"2026-07-05T11:14:54.396869+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.08231","citing_title":"Test-Time Scaling in Multimodal Foundation Models: A Comprehensive Survey of Generation and Reasoning","ref_index":80,"is_internal_anchor":false},{"citing_arxiv_id":"2601.05110","citing_title":"GlimpRouter: Efficient Collaborative Inference by Glimpsing One Token of Thoughts","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2504.19678","citing_title":"From LLM Reasoning to Autonomous AI Agents: A Comprehensive Review","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10167","citing_title":"Visual Late Chunking: An Empirical Study of Contextual Chunking for Efficient Visual Document Retrieval","ref_index":39,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/L6RYV4JJSLKKB4EMRZM6PU3ZHU","json":"https://pith.science/pith/L6RYV4JJSLKKB4EMRZM6PU3ZHU.json","graph_json":"https://pith.science/api/pith-number/L6RYV4JJSLKKB4EMRZM6PU3ZHU/graph.json","events_json":"https://pith.science/api/pith-number/L6RYV4JJSLKKB4EMRZM6PU3ZHU/events.json","paper":"https://pith.science/paper/L6RYV4JJ"},"agent_actions":{"view_html":"https://pith.science/pith/L6RYV4JJSLKKB4EMRZM6PU3ZHU","download_json":"https://pith.science/pith/L6RYV4JJSLKKB4EMRZM6PU3ZHU.json","view_paper":"https://pith.science/paper/L6RYV4JJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.18017&json=true","fetch_graph":"https://pith.science/api/pith-number/L6RYV4JJSLKKB4EMRZM6PU3ZHU/graph.json","fetch_events":"https://pith.science/api/pith-number/L6RYV4JJSLKKB4EMRZM6PU3ZHU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/L6RYV4JJSLKKB4EMRZM6PU3ZHU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/L6RYV4JJSLKKB4EMRZM6PU3ZHU/action/storage_attestation","attest_author":"https://pith.science/pith/L6RYV4JJSLKKB4EMRZM6PU3ZHU/action/author_attestation","sign_citation":"https://pith.science/pith/L6RYV4JJSLKKB4EMRZM6PU3ZHU/action/citation_signature","submit_replication":"https://pith.science/pith/L6RYV4JJSLKKB4EMRZM6PU3ZHU/action/replication_record"}},"created_at":"2026-07-05T11:14:54.396869+00:00","updated_at":"2026-07-05T11:14:54.396869+00:00"}