{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:PZPCGIUD6O7FDACUCVCE6YICVN","short_pith_number":"pith:PZPCGIUD","schema_version":"1.0","canonical_sha256":"7e5e232283f3be51805415444f6102ab7240697ca37cf0dc5744c41eaafb01f0","source":{"kind":"arxiv","id":"2505.24073","version":2},"attestation_state":"computed","paper":{"title":"mRAG: Elucidating the Design Space of Multi-modal Retrieval-Augmented Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.CV"],"primary_cat":"cs.AI","authors_text":"Chan-Wei Hu, Chia-Ju Chen, Ryan Rossi, Shuo Xing, Suofei Feng, Yueqi Wang, Zhengzhong Tu","submitted_at":"2025-05-29T23:32:03Z","abstract_excerpt":"Large Vision-Language Models (LVLMs) have made remarkable strides in multimodal tasks such as visual question answering, visual grounding, and complex reasoning. However, they remain limited by static training data, susceptibility to hallucinations, and inability to verify claims against up-to-date, external evidence, compromising their performance in dynamic real-world applications. Retrieval-Augmented Generation (RAG) offers a practical solution to mitigate these challenges by allowing the LVLMs to access large-scale knowledge databases via retrieval mechanisms, thereby grounding model outpu"},"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":"2505.24073","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-29T23:32:03Z","cross_cats_sorted":["cs.CL","cs.CV"],"title_canon_sha256":"65e3ea565ff08fc8883c05f473022c9c760fd3b5eecf360009b151a46c1c296b","abstract_canon_sha256":"d7d6589d8dcd93c6716b12277b1367446d27a2d485da4387e2ac00558cabe388"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:59:15.214854Z","signature_b64":"0KE8PfEIezqIQYQeyq8FF+nD0fPuWeVH2Bggv94Jr5rgW3++iaTHbY2dO5yzoDXtOgQbOo06zrMb5cgw98osCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7e5e232283f3be51805415444f6102ab7240697ca37cf0dc5744c41eaafb01f0","last_reissued_at":"2026-07-05T11:59:15.214417Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:59:15.214417Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"mRAG: Elucidating the Design Space of Multi-modal Retrieval-Augmented Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.CV"],"primary_cat":"cs.AI","authors_text":"Chan-Wei Hu, Chia-Ju Chen, Ryan Rossi, Shuo Xing, Suofei Feng, Yueqi Wang, Zhengzhong Tu","submitted_at":"2025-05-29T23:32:03Z","abstract_excerpt":"Large Vision-Language Models (LVLMs) have made remarkable strides in multimodal tasks such as visual question answering, visual grounding, and complex reasoning. However, they remain limited by static training data, susceptibility to hallucinations, and inability to verify claims against up-to-date, external evidence, compromising their performance in dynamic real-world applications. Retrieval-Augmented Generation (RAG) offers a practical solution to mitigate these challenges by allowing the LVLMs to access large-scale knowledge databases via retrieval mechanisms, thereby grounding model outpu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.24073","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/2505.24073/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":"2505.24073","created_at":"2026-07-05T11:59:15.214476+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.24073v2","created_at":"2026-07-05T11:59:15.214476+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.24073","created_at":"2026-07-05T11:59:15.214476+00:00"},{"alias_kind":"pith_short_12","alias_value":"PZPCGIUD6O7F","created_at":"2026-07-05T11:59:15.214476+00:00"},{"alias_kind":"pith_short_16","alias_value":"PZPCGIUD6O7FDACU","created_at":"2026-07-05T11:59:15.214476+00:00"},{"alias_kind":"pith_short_8","alias_value":"PZPCGIUD","created_at":"2026-07-05T11:59:15.214476+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24107","citing_title":"DramaDirector: Geometry-Guided Short Drama Generation","ref_index":50,"is_internal_anchor":false},{"citing_arxiv_id":"2606.16494","citing_title":"Lost at the End: Primacy Bias in Multimodal Retrieval-Augmented Question Answering","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04969","citing_title":"MG$^2$-RAG: Multi-Granularity Graph for Multimodal Retrieval-Augmented Generation","ref_index":31,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PZPCGIUD6O7FDACUCVCE6YICVN","json":"https://pith.science/pith/PZPCGIUD6O7FDACUCVCE6YICVN.json","graph_json":"https://pith.science/api/pith-number/PZPCGIUD6O7FDACUCVCE6YICVN/graph.json","events_json":"https://pith.science/api/pith-number/PZPCGIUD6O7FDACUCVCE6YICVN/events.json","paper":"https://pith.science/paper/PZPCGIUD"},"agent_actions":{"view_html":"https://pith.science/pith/PZPCGIUD6O7FDACUCVCE6YICVN","download_json":"https://pith.science/pith/PZPCGIUD6O7FDACUCVCE6YICVN.json","view_paper":"https://pith.science/paper/PZPCGIUD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.24073&json=true","fetch_graph":"https://pith.science/api/pith-number/PZPCGIUD6O7FDACUCVCE6YICVN/graph.json","fetch_events":"https://pith.science/api/pith-number/PZPCGIUD6O7FDACUCVCE6YICVN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PZPCGIUD6O7FDACUCVCE6YICVN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PZPCGIUD6O7FDACUCVCE6YICVN/action/storage_attestation","attest_author":"https://pith.science/pith/PZPCGIUD6O7FDACUCVCE6YICVN/action/author_attestation","sign_citation":"https://pith.science/pith/PZPCGIUD6O7FDACUCVCE6YICVN/action/citation_signature","submit_replication":"https://pith.science/pith/PZPCGIUD6O7FDACUCVCE6YICVN/action/replication_record"}},"created_at":"2026-07-05T11:59:15.214476+00:00","updated_at":"2026-07-05T11:59:15.214476+00:00"}