{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:LKU7MH5WJYGKTYBEB2TS2YFILR","short_pith_number":"pith:LKU7MH5W","schema_version":"1.0","canonical_sha256":"5aa9f61fb64e0ca9e0240ea72d60a85c7f1d4d7f205d9409736e60a59cf1836f","source":{"kind":"arxiv","id":"2607.20517","version":1},"attestation_state":"computed","paper":{"title":"Multimodal CoLRAG-TF: Triple-Filtered Retrieval for Complex PDFs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Takato Yasuno","submitted_at":"2026-07-07T23:00:52Z","abstract_excerpt":"Retrieval-augmented generation (RAG) over heterogeneous PDF collections remains challenging due to multimodal content, domain-specific terminology, and the need for multi-hop reasoning across dispersed evidence. We present Multimodal CoLRAG-TF, a four-axis fusion architecture that integrates dense text embeddings, BM25 keyword matching, knowledge-graph triple filtering, and image-based similarity for robust retrieval over complex documents. Our system constructs a multimodal index of 2,403 blocks extracted from 43 Japanese disaster lesson PDFs, supported by a hybrid OCR pipeline and LLM-based "},"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":"2607.20517","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-07-07T23:00:52Z","cross_cats_sorted":[],"title_canon_sha256":"e2174a6b6ccbdd654cebdc2d9954ac52ad43325415580b72992166f1a92d02f6","abstract_canon_sha256":"3e2f9c57a7e017b9381b1f30b4e6e6a9f9d621a68a3c02477bf3cfc4828a9607"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-24T00:23:19.951917Z","signature_b64":"M2YpoAFJWe0gdqS9zScORRCQQa1wlIPxhrLiHH7quhraprMQMvq+6RJ8EDWzgtIaUJa4GzYCYSaR0uHxihOKCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5aa9f61fb64e0ca9e0240ea72d60a85c7f1d4d7f205d9409736e60a59cf1836f","last_reissued_at":"2026-07-24T00:23:19.951043Z","signature_status":"signed_v1","first_computed_at":"2026-07-24T00:23:19.951043Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multimodal CoLRAG-TF: Triple-Filtered Retrieval for Complex PDFs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Takato Yasuno","submitted_at":"2026-07-07T23:00:52Z","abstract_excerpt":"Retrieval-augmented generation (RAG) over heterogeneous PDF collections remains challenging due to multimodal content, domain-specific terminology, and the need for multi-hop reasoning across dispersed evidence. We present Multimodal CoLRAG-TF, a four-axis fusion architecture that integrates dense text embeddings, BM25 keyword matching, knowledge-graph triple filtering, and image-based similarity for robust retrieval over complex documents. Our system constructs a multimodal index of 2,403 blocks extracted from 43 Japanese disaster lesson PDFs, supported by a hybrid OCR pipeline and LLM-based "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.20517","kind":"arxiv","version":1},"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/2607.20517/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":"2607.20517","created_at":"2026-07-24T00:23:19.951508+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.20517v1","created_at":"2026-07-24T00:23:19.951508+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.20517","created_at":"2026-07-24T00:23:19.951508+00:00"},{"alias_kind":"pith_short_12","alias_value":"LKU7MH5WJYGK","created_at":"2026-07-24T00:23:19.951508+00:00"},{"alias_kind":"pith_short_16","alias_value":"LKU7MH5WJYGKTYBE","created_at":"2026-07-24T00:23:19.951508+00:00"},{"alias_kind":"pith_short_8","alias_value":"LKU7MH5W","created_at":"2026-07-24T00:23:19.951508+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LKU7MH5WJYGKTYBEB2TS2YFILR","json":"https://pith.science/pith/LKU7MH5WJYGKTYBEB2TS2YFILR.json","graph_json":"https://pith.science/api/pith-number/LKU7MH5WJYGKTYBEB2TS2YFILR/graph.json","events_json":"https://pith.science/api/pith-number/LKU7MH5WJYGKTYBEB2TS2YFILR/events.json","paper":"https://pith.science/paper/LKU7MH5W"},"agent_actions":{"view_html":"https://pith.science/pith/LKU7MH5WJYGKTYBEB2TS2YFILR","download_json":"https://pith.science/pith/LKU7MH5WJYGKTYBEB2TS2YFILR.json","view_paper":"https://pith.science/paper/LKU7MH5W","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.20517&json=true","fetch_graph":"https://pith.science/api/pith-number/LKU7MH5WJYGKTYBEB2TS2YFILR/graph.json","fetch_events":"https://pith.science/api/pith-number/LKU7MH5WJYGKTYBEB2TS2YFILR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LKU7MH5WJYGKTYBEB2TS2YFILR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LKU7MH5WJYGKTYBEB2TS2YFILR/action/storage_attestation","attest_author":"https://pith.science/pith/LKU7MH5WJYGKTYBEB2TS2YFILR/action/author_attestation","sign_citation":"https://pith.science/pith/LKU7MH5WJYGKTYBEB2TS2YFILR/action/citation_signature","submit_replication":"https://pith.science/pith/LKU7MH5WJYGKTYBEB2TS2YFILR/action/replication_record"}},"created_at":"2026-07-24T00:23:19.951508+00:00","updated_at":"2026-07-24T00:23:19.951508+00:00"}