{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:ENLCXHM4Z4MVYCWW3GZ347DMS4","short_pith_number":"pith:ENLCXHM4","schema_version":"1.0","canonical_sha256":"23562b9d9ccf195c0ad6d9b3be7c6c97039a91a73aa59a931a5447ceb2652ed9","source":{"kind":"arxiv","id":"2406.12746","version":5},"attestation_state":"computed","paper":{"title":"Diversify, Rationalize, and Combine: Ensembling Multiple QA Strategies for Zero-shot Knowledge-based VQA","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Boyang Li, Haoxin Li, Miaoyu Li, Zilin Du","submitted_at":"2024-06-18T16:06:38Z","abstract_excerpt":"Knowledge-based Visual Question-answering (K-VQA) often requires the use of background knowledge beyond the image. However, we discover that a single knowledge generation strategy is often insufficient for all K-VQA questions. To this end, we propose Diversification, Evidence Truncation, and Combination for Knowledge-based Elucidation (DietCoke), which utilizes a bundle of complementary question-answering tactics and aggregates their answers using textual rationales. DietCoke comprises of three stages: diversification, rationalization, and ensemble. The diversification stage generates three di"},"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":"2406.12746","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-18T16:06:38Z","cross_cats_sorted":[],"title_canon_sha256":"34d03371c39f93da7e0a6332c71cf21e3a6029f6868e8778e3c88a6fd4cd9e1b","abstract_canon_sha256":"5f0cbf3faa289893ac7ec71a046f10a372da2d5e0aa04753ec19ae72e22da7a2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:17:43.573330Z","signature_b64":"Rv9XkG83ilTNfWLI0GSZaWKjFtwbvXpN2nNiW87YovKaimWLstmrMpwcVYOsqw+smgifBF1MmS4LVC3KrBC0BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"23562b9d9ccf195c0ad6d9b3be7c6c97039a91a73aa59a931a5447ceb2652ed9","last_reissued_at":"2026-07-05T09:17:43.572850Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:17:43.572850Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Diversify, Rationalize, and Combine: Ensembling Multiple QA Strategies for Zero-shot Knowledge-based VQA","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Boyang Li, Haoxin Li, Miaoyu Li, Zilin Du","submitted_at":"2024-06-18T16:06:38Z","abstract_excerpt":"Knowledge-based Visual Question-answering (K-VQA) often requires the use of background knowledge beyond the image. However, we discover that a single knowledge generation strategy is often insufficient for all K-VQA questions. To this end, we propose Diversification, Evidence Truncation, and Combination for Knowledge-based Elucidation (DietCoke), which utilizes a bundle of complementary question-answering tactics and aggregates their answers using textual rationales. DietCoke comprises of three stages: diversification, rationalization, and ensemble. The diversification stage generates three di"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.12746","kind":"arxiv","version":5},"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/2406.12746/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":"2406.12746","created_at":"2026-07-05T09:17:43.572903+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.12746v5","created_at":"2026-07-05T09:17:43.572903+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.12746","created_at":"2026-07-05T09:17:43.572903+00:00"},{"alias_kind":"pith_short_12","alias_value":"ENLCXHM4Z4MV","created_at":"2026-07-05T09:17:43.572903+00:00"},{"alias_kind":"pith_short_16","alias_value":"ENLCXHM4Z4MVYCWW","created_at":"2026-07-05T09:17:43.572903+00:00"},{"alias_kind":"pith_short_8","alias_value":"ENLCXHM4","created_at":"2026-07-05T09:17:43.572903+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.00711","citing_title":"VIKSER: Visual Knowledge-Driven Self-Reinforcing Reasoning Framework","ref_index":48,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ENLCXHM4Z4MVYCWW3GZ347DMS4","json":"https://pith.science/pith/ENLCXHM4Z4MVYCWW3GZ347DMS4.json","graph_json":"https://pith.science/api/pith-number/ENLCXHM4Z4MVYCWW3GZ347DMS4/graph.json","events_json":"https://pith.science/api/pith-number/ENLCXHM4Z4MVYCWW3GZ347DMS4/events.json","paper":"https://pith.science/paper/ENLCXHM4"},"agent_actions":{"view_html":"https://pith.science/pith/ENLCXHM4Z4MVYCWW3GZ347DMS4","download_json":"https://pith.science/pith/ENLCXHM4Z4MVYCWW3GZ347DMS4.json","view_paper":"https://pith.science/paper/ENLCXHM4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.12746&json=true","fetch_graph":"https://pith.science/api/pith-number/ENLCXHM4Z4MVYCWW3GZ347DMS4/graph.json","fetch_events":"https://pith.science/api/pith-number/ENLCXHM4Z4MVYCWW3GZ347DMS4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ENLCXHM4Z4MVYCWW3GZ347DMS4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ENLCXHM4Z4MVYCWW3GZ347DMS4/action/storage_attestation","attest_author":"https://pith.science/pith/ENLCXHM4Z4MVYCWW3GZ347DMS4/action/author_attestation","sign_citation":"https://pith.science/pith/ENLCXHM4Z4MVYCWW3GZ347DMS4/action/citation_signature","submit_replication":"https://pith.science/pith/ENLCXHM4Z4MVYCWW3GZ347DMS4/action/replication_record"}},"created_at":"2026-07-05T09:17:43.572903+00:00","updated_at":"2026-07-05T09:17:43.572903+00:00"}