{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:4PLA2XMAEZLNEXL2JFTJSIBT3J","short_pith_number":"pith:4PLA2XMA","schema_version":"1.0","canonical_sha256":"e3d60d5d802656d25d7a4966992033da4c359fe0fbe33397b4d034e9e46c1119","source":{"kind":"arxiv","id":"2402.08327","version":2},"attestation_state":"computed","paper":{"title":"PreFLMR: Scaling Up Fine-Grained Late-Interaction Multi-modal Retrievers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bill Byrne, Jingbiao Mei, Jinghong Chen, Weizhe Lin","submitted_at":"2024-02-13T09:47:07Z","abstract_excerpt":"Large Multimodal Models (LMMs) excel in natural language and visual understanding but are challenged by exacting tasks such as Knowledge-based Visual Question Answering (KB-VQA) which involve the retrieval of relevant information from document collections to use in shaping answers to questions. We present an extensive training and evaluation framework, M2KR, for KB-VQA. M2KR contains a collection of vision and language tasks which we have incorporated into a single suite of benchmark tasks for training and evaluating general-purpose multi-modal retrievers. We use M2KR to develop PreFLMR, a pre"},"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":"2402.08327","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-13T09:47:07Z","cross_cats_sorted":[],"title_canon_sha256":"e43c6a23a1e1d144103afb995e7b8189086f09de08ed08db4fb8101be9c19201","abstract_canon_sha256":"fd1f0949b20f7ac1345cbfd7e6cdaf99e8e1285263972d833cac9517e3f087b9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:27:45.989160Z","signature_b64":"S8qqg9iplCwk1bFOA5bYscUe6bQBqAxyVZIW6ZEcWvomygD1ujy7SV9FDt7SKo/oToj82nXpDqyVfSCCdm8oBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e3d60d5d802656d25d7a4966992033da4c359fe0fbe33397b4d034e9e46c1119","last_reissued_at":"2026-07-05T08:27:45.988610Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:27:45.988610Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PreFLMR: Scaling Up Fine-Grained Late-Interaction Multi-modal Retrievers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bill Byrne, Jingbiao Mei, Jinghong Chen, Weizhe Lin","submitted_at":"2024-02-13T09:47:07Z","abstract_excerpt":"Large Multimodal Models (LMMs) excel in natural language and visual understanding but are challenged by exacting tasks such as Knowledge-based Visual Question Answering (KB-VQA) which involve the retrieval of relevant information from document collections to use in shaping answers to questions. We present an extensive training and evaluation framework, M2KR, for KB-VQA. M2KR contains a collection of vision and language tasks which we have incorporated into a single suite of benchmark tasks for training and evaluating general-purpose multi-modal retrievers. We use M2KR to develop PreFLMR, a pre"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.08327","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/2402.08327/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":"2402.08327","created_at":"2026-07-05T08:27:45.988672+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.08327v2","created_at":"2026-07-05T08:27:45.988672+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.08327","created_at":"2026-07-05T08:27:45.988672+00:00"},{"alias_kind":"pith_short_12","alias_value":"4PLA2XMAEZLN","created_at":"2026-07-05T08:27:45.988672+00:00"},{"alias_kind":"pith_short_16","alias_value":"4PLA2XMAEZLNEXL2","created_at":"2026-07-05T08:27:45.988672+00:00"},{"alias_kind":"pith_short_8","alias_value":"4PLA2XMA","created_at":"2026-07-05T08:27:45.988672+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.23881","citing_title":"Ground Then Rank: Revisiting Knowledge-Based VQA with Training-Free Entity Identification","ref_index":30,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4PLA2XMAEZLNEXL2JFTJSIBT3J","json":"https://pith.science/pith/4PLA2XMAEZLNEXL2JFTJSIBT3J.json","graph_json":"https://pith.science/api/pith-number/4PLA2XMAEZLNEXL2JFTJSIBT3J/graph.json","events_json":"https://pith.science/api/pith-number/4PLA2XMAEZLNEXL2JFTJSIBT3J/events.json","paper":"https://pith.science/paper/4PLA2XMA"},"agent_actions":{"view_html":"https://pith.science/pith/4PLA2XMAEZLNEXL2JFTJSIBT3J","download_json":"https://pith.science/pith/4PLA2XMAEZLNEXL2JFTJSIBT3J.json","view_paper":"https://pith.science/paper/4PLA2XMA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.08327&json=true","fetch_graph":"https://pith.science/api/pith-number/4PLA2XMAEZLNEXL2JFTJSIBT3J/graph.json","fetch_events":"https://pith.science/api/pith-number/4PLA2XMAEZLNEXL2JFTJSIBT3J/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4PLA2XMAEZLNEXL2JFTJSIBT3J/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4PLA2XMAEZLNEXL2JFTJSIBT3J/action/storage_attestation","attest_author":"https://pith.science/pith/4PLA2XMAEZLNEXL2JFTJSIBT3J/action/author_attestation","sign_citation":"https://pith.science/pith/4PLA2XMAEZLNEXL2JFTJSIBT3J/action/citation_signature","submit_replication":"https://pith.science/pith/4PLA2XMAEZLNEXL2JFTJSIBT3J/action/replication_record"}},"created_at":"2026-07-05T08:27:45.988672+00:00","updated_at":"2026-07-05T08:27:45.988672+00:00"}