{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:MG54F3ZQSYHY437LSVHERQEF5J","short_pith_number":"pith:MG54F3ZQ","schema_version":"1.0","canonical_sha256":"61bbc2ef30960f8e6feb954e48c085ea63f37ec0a6cb449b2015c4fa26627f1b","source":{"kind":"arxiv","id":"2411.01106","version":2},"attestation_state":"computed","paper":{"title":"SV-RAG: LoRA-Contextualizing Adaptation of MLLMs for Long Document Understanding","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Changyou Chen, Franck Dernoncourt, Jian Chen, Jiuxiang Gu, Ruiyi Zhang, Ryan A. Rossi, Tong Sun, Tong Yu, Yufan Zhou","submitted_at":"2024-11-02T02:09:01Z","abstract_excerpt":"Multimodal large language models (MLLMs) have recently shown great progress in text-rich image understanding, yet they still struggle with complex, multi-page visually-rich documents. Traditional methods using document parsers for retrieval-augmented generation suffer from performance and efficiency limitations, while directly presenting all pages to MLLMs leads to inefficiencies, especially with lengthy ones. In this work, we present a novel framework named **S**elf-**V**isual **R**etrieval-**A**ugmented **G**eneration (SV-RAG), which can broaden horizons of any MLLM to support long-document "},"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":"2411.01106","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2024-11-02T02:09:01Z","cross_cats_sorted":[],"title_canon_sha256":"b55e664516d6ce8a886b7946f8313e525d24465763e92fb4a57b814c0ca87595","abstract_canon_sha256":"9313da74f8c97a8c05d3bcbc90eb851bfd55f185e84cb3e933be46d511cc6a45"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:22:09.030502Z","signature_b64":"HFA03T12Z8EVkcEOTYDA1u5qXHKLp6O7TU5sdT8Dr/rrzKnfI1xEUPy5RA5dOFhfepcozv0wYEYgCRD5iRn2Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"61bbc2ef30960f8e6feb954e48c085ea63f37ec0a6cb449b2015c4fa26627f1b","last_reissued_at":"2026-07-05T10:22:09.029994Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:22:09.029994Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SV-RAG: LoRA-Contextualizing Adaptation of MLLMs for Long Document Understanding","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Changyou Chen, Franck Dernoncourt, Jian Chen, Jiuxiang Gu, Ruiyi Zhang, Ryan A. Rossi, Tong Sun, Tong Yu, Yufan Zhou","submitted_at":"2024-11-02T02:09:01Z","abstract_excerpt":"Multimodal large language models (MLLMs) have recently shown great progress in text-rich image understanding, yet they still struggle with complex, multi-page visually-rich documents. Traditional methods using document parsers for retrieval-augmented generation suffer from performance and efficiency limitations, while directly presenting all pages to MLLMs leads to inefficiencies, especially with lengthy ones. In this work, we present a novel framework named **S**elf-**V**isual **R**etrieval-**A**ugmented **G**eneration (SV-RAG), which can broaden horizons of any MLLM to support long-document "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.01106","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/2411.01106/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":"2411.01106","created_at":"2026-07-05T10:22:09.030058+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.01106v2","created_at":"2026-07-05T10:22:09.030058+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.01106","created_at":"2026-07-05T10:22:09.030058+00:00"},{"alias_kind":"pith_short_12","alias_value":"MG54F3ZQSYHY","created_at":"2026-07-05T10:22:09.030058+00:00"},{"alias_kind":"pith_short_16","alias_value":"MG54F3ZQSYHY437L","created_at":"2026-07-05T10:22:09.030058+00:00"},{"alias_kind":"pith_short_8","alias_value":"MG54F3ZQ","created_at":"2026-07-05T10:22:09.030058+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2602.23061","citing_title":"MoDora: Tree-Based Semi-Structured Document Analysis System","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12812","citing_title":"DocSeeker: Structured Visual Reasoning with Evidence Grounding for Long Document Understanding","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12812","citing_title":"DocSeeker: Structured Visual Reasoning with Evidence Grounding for Long Document Understanding","ref_index":32,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MG54F3ZQSYHY437LSVHERQEF5J","json":"https://pith.science/pith/MG54F3ZQSYHY437LSVHERQEF5J.json","graph_json":"https://pith.science/api/pith-number/MG54F3ZQSYHY437LSVHERQEF5J/graph.json","events_json":"https://pith.science/api/pith-number/MG54F3ZQSYHY437LSVHERQEF5J/events.json","paper":"https://pith.science/paper/MG54F3ZQ"},"agent_actions":{"view_html":"https://pith.science/pith/MG54F3ZQSYHY437LSVHERQEF5J","download_json":"https://pith.science/pith/MG54F3ZQSYHY437LSVHERQEF5J.json","view_paper":"https://pith.science/paper/MG54F3ZQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.01106&json=true","fetch_graph":"https://pith.science/api/pith-number/MG54F3ZQSYHY437LSVHERQEF5J/graph.json","fetch_events":"https://pith.science/api/pith-number/MG54F3ZQSYHY437LSVHERQEF5J/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MG54F3ZQSYHY437LSVHERQEF5J/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MG54F3ZQSYHY437LSVHERQEF5J/action/storage_attestation","attest_author":"https://pith.science/pith/MG54F3ZQSYHY437LSVHERQEF5J/action/author_attestation","sign_citation":"https://pith.science/pith/MG54F3ZQSYHY437LSVHERQEF5J/action/citation_signature","submit_replication":"https://pith.science/pith/MG54F3ZQSYHY437LSVHERQEF5J/action/replication_record"}},"created_at":"2026-07-05T10:22:09.030058+00:00","updated_at":"2026-07-05T10:22:09.030058+00:00"}