{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:6MSFQPEMYVS56RL7IZRXMDZRTN","short_pith_number":"pith:6MSFQPEM","schema_version":"1.0","canonical_sha256":"f324583c8cc565df457f4663760f319b609b1710c34167b0884ed42b91a91527","source":{"kind":"arxiv","id":"2508.18984","version":2},"attestation_state":"computed","paper":{"title":"Enhancing Document VQA Models via Retrieval-Augmented Generation","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Artemis LLabr\\'es, Eric L\\'opez, Ernest Valveny","submitted_at":"2025-08-26T12:32:55Z","abstract_excerpt":"Document Visual Question Answering (Document VQA) must cope with documents that span dozens of pages, yet leading systems still concatenate every page or rely on very large vision-language models, both of which are memory-hungry. Retrieval-Augmented Generation (RAG) offers an attractive alternative, first retrieving a concise set of relevant segments before generating answers from this selected evidence. In this paper, we systematically evaluate the impact of incorporating RAG into Document VQA through different retrieval variants - text-based retrieval using OCR tokens and purely visual retri"},"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":"2508.18984","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-08-26T12:32:55Z","cross_cats_sorted":[],"title_canon_sha256":"e53f5d8493f6b022203d875d50ed2d18df0248ca7ce148fa531757ec15d85c2d","abstract_canon_sha256":"31163ced39864e66d7d8782afb3587bce807ff1ebfaaa60ae709d66ec3dd49a3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:00:52.171420Z","signature_b64":"M2DRZU6KGFNSkZsrFWBRyCZltjpf9JqkYjDnDOXL9uk60TkbZjP3s4YpyLE9b2kENXhejRsr700ntZdv5ATLBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f324583c8cc565df457f4663760f319b609b1710c34167b0884ed42b91a91527","last_reissued_at":"2026-07-05T12:00:52.170855Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:00:52.170855Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Enhancing Document VQA Models via Retrieval-Augmented Generation","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Artemis LLabr\\'es, Eric L\\'opez, Ernest Valveny","submitted_at":"2025-08-26T12:32:55Z","abstract_excerpt":"Document Visual Question Answering (Document VQA) must cope with documents that span dozens of pages, yet leading systems still concatenate every page or rely on very large vision-language models, both of which are memory-hungry. Retrieval-Augmented Generation (RAG) offers an attractive alternative, first retrieving a concise set of relevant segments before generating answers from this selected evidence. In this paper, we systematically evaluate the impact of incorporating RAG into Document VQA through different retrieval variants - text-based retrieval using OCR tokens and purely visual retri"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.18984","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/2508.18984/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":"2508.18984","created_at":"2026-07-05T12:00:52.170923+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.18984v2","created_at":"2026-07-05T12:00:52.170923+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.18984","created_at":"2026-07-05T12:00:52.170923+00:00"},{"alias_kind":"pith_short_12","alias_value":"6MSFQPEMYVS5","created_at":"2026-07-05T12:00:52.170923+00:00"},{"alias_kind":"pith_short_16","alias_value":"6MSFQPEMYVS56RL7","created_at":"2026-07-05T12:00:52.170923+00:00"},{"alias_kind":"pith_short_8","alias_value":"6MSFQPEM","created_at":"2026-07-05T12:00:52.170923+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2510.15253","citing_title":"Scaling Beyond Context: A Survey of Multimodal Retrieval-Augmented Generation for Document Understanding","ref_index":4,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6MSFQPEMYVS56RL7IZRXMDZRTN","json":"https://pith.science/pith/6MSFQPEMYVS56RL7IZRXMDZRTN.json","graph_json":"https://pith.science/api/pith-number/6MSFQPEMYVS56RL7IZRXMDZRTN/graph.json","events_json":"https://pith.science/api/pith-number/6MSFQPEMYVS56RL7IZRXMDZRTN/events.json","paper":"https://pith.science/paper/6MSFQPEM"},"agent_actions":{"view_html":"https://pith.science/pith/6MSFQPEMYVS56RL7IZRXMDZRTN","download_json":"https://pith.science/pith/6MSFQPEMYVS56RL7IZRXMDZRTN.json","view_paper":"https://pith.science/paper/6MSFQPEM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.18984&json=true","fetch_graph":"https://pith.science/api/pith-number/6MSFQPEMYVS56RL7IZRXMDZRTN/graph.json","fetch_events":"https://pith.science/api/pith-number/6MSFQPEMYVS56RL7IZRXMDZRTN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6MSFQPEMYVS56RL7IZRXMDZRTN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6MSFQPEMYVS56RL7IZRXMDZRTN/action/storage_attestation","attest_author":"https://pith.science/pith/6MSFQPEMYVS56RL7IZRXMDZRTN/action/author_attestation","sign_citation":"https://pith.science/pith/6MSFQPEMYVS56RL7IZRXMDZRTN/action/citation_signature","submit_replication":"https://pith.science/pith/6MSFQPEMYVS56RL7IZRXMDZRTN/action/replication_record"}},"created_at":"2026-07-05T12:00:52.170923+00:00","updated_at":"2026-07-05T12:00:52.170923+00:00"}