{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:7TV6CLXO2IB7IOULHKRYJE5AG7","short_pith_number":"pith:7TV6CLXO","schema_version":"1.0","canonical_sha256":"fcebe12eeed203f43a8b3aa38493a037e6e2be4f06d81e97d94d9aff69b165ba","source":{"kind":"arxiv","id":"2608.02907","version":1},"attestation_state":"computed","paper":{"title":"Bayesian Data Reweighting Improves Multimodal Retrieval for Knowledge-Based Visual Question Answering","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Changyou Chen, Jingchen Sun, Ming Liu, Naresh Kumar Devulapally, Ruiyi Zhang, Shaobo Han, Vishnu Suresh Lokhande, Yitao Long","submitted_at":"2026-08-03T21:46:08Z","abstract_excerpt":"Multimodal retrievers are essential for knowledge-based visual question answering, where they retrieve external evidence for image-question pairs. However, existing contrastive training methods typically treat all unmatched query-document pairs as equally informative negatives, which is problematic because many unmatched documents may still be semantically relevant or partially useful. We propose Bayesian Data Reweighting, a probabilistic framework that models query-document importance as latent variables and adaptively infers posterior weights to downweight likely false negatives. With closed"},"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":"2608.02907","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2026-08-03T21:46:08Z","cross_cats_sorted":[],"title_canon_sha256":"6ec3d0105508df8adafb25081d944329e70c502a97f67806343dbe7f2899a874","abstract_canon_sha256":"1f7d804247e31214d83f0c04871a690a0b620f229f22093adeb1ef222fc4987e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-05T00:43:35.607037Z","signature_b64":"UF2XVyy4/B0Wk0x5dxlbn75ejThfm6U5dGvZxBd9fjli4vQDbX9zAFq8aom5xPx6/jq6vNW0no8ZP/ucQXWhBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fcebe12eeed203f43a8b3aa38493a037e6e2be4f06d81e97d94d9aff69b165ba","last_reissued_at":"2026-08-05T00:43:35.604597Z","signature_status":"signed_v1","first_computed_at":"2026-08-05T00:43:35.604597Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Bayesian Data Reweighting Improves Multimodal Retrieval for Knowledge-Based Visual Question Answering","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Changyou Chen, Jingchen Sun, Ming Liu, Naresh Kumar Devulapally, Ruiyi Zhang, Shaobo Han, Vishnu Suresh Lokhande, Yitao Long","submitted_at":"2026-08-03T21:46:08Z","abstract_excerpt":"Multimodal retrievers are essential for knowledge-based visual question answering, where they retrieve external evidence for image-question pairs. However, existing contrastive training methods typically treat all unmatched query-document pairs as equally informative negatives, which is problematic because many unmatched documents may still be semantically relevant or partially useful. We propose Bayesian Data Reweighting, a probabilistic framework that models query-document importance as latent variables and adaptively infers posterior weights to downweight likely false negatives. With closed"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.02907","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/2608.02907/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":"2608.02907","created_at":"2026-08-05T00:43:35.605558+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.02907v1","created_at":"2026-08-05T00:43:35.605558+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.02907","created_at":"2026-08-05T00:43:35.605558+00:00"},{"alias_kind":"pith_short_12","alias_value":"7TV6CLXO2IB7","created_at":"2026-08-05T00:43:35.605558+00:00"},{"alias_kind":"pith_short_16","alias_value":"7TV6CLXO2IB7IOUL","created_at":"2026-08-05T00:43:35.605558+00:00"},{"alias_kind":"pith_short_8","alias_value":"7TV6CLXO","created_at":"2026-08-05T00:43:35.605558+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/7TV6CLXO2IB7IOULHKRYJE5AG7","json":"https://pith.science/pith/7TV6CLXO2IB7IOULHKRYJE5AG7.json","graph_json":"https://pith.science/api/pith-number/7TV6CLXO2IB7IOULHKRYJE5AG7/graph.json","events_json":"https://pith.science/api/pith-number/7TV6CLXO2IB7IOULHKRYJE5AG7/events.json","paper":"https://pith.science/paper/7TV6CLXO"},"agent_actions":{"view_html":"https://pith.science/pith/7TV6CLXO2IB7IOULHKRYJE5AG7","download_json":"https://pith.science/pith/7TV6CLXO2IB7IOULHKRYJE5AG7.json","view_paper":"https://pith.science/paper/7TV6CLXO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.02907&json=true","fetch_graph":"https://pith.science/api/pith-number/7TV6CLXO2IB7IOULHKRYJE5AG7/graph.json","fetch_events":"https://pith.science/api/pith-number/7TV6CLXO2IB7IOULHKRYJE5AG7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7TV6CLXO2IB7IOULHKRYJE5AG7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7TV6CLXO2IB7IOULHKRYJE5AG7/action/storage_attestation","attest_author":"https://pith.science/pith/7TV6CLXO2IB7IOULHKRYJE5AG7/action/author_attestation","sign_citation":"https://pith.science/pith/7TV6CLXO2IB7IOULHKRYJE5AG7/action/citation_signature","submit_replication":"https://pith.science/pith/7TV6CLXO2IB7IOULHKRYJE5AG7/action/replication_record"}},"created_at":"2026-08-05T00:43:35.605558+00:00","updated_at":"2026-08-05T00:43:35.605558+00:00"}