{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:RU2YHXYV3WXTANXGAMGM7O7KGU","short_pith_number":"pith:RU2YHXYV","schema_version":"1.0","canonical_sha256":"8d3583df15ddaf3036e6030ccfbbea3511ed55c272e10ef5f91e1f90f56dc8c3","source":{"kind":"arxiv","id":"2502.20964","version":3},"attestation_state":"computed","paper":{"title":"Fine-Grained Knowledge Structuring and Retrieval for Visual Question Answering","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Nan Tang, Yin Wu, Yuyu Luo, Zhengxuan Zhang","submitted_at":"2025-02-28T11:25:38Z","abstract_excerpt":"Visual Question Answering (VQA) focuses on providing answers to natural language questions by utilizing information from images. Although cutting-edge multimodal large language models (MLLMs) such as GPT-4o achieve strong performance on VQA tasks, they frequently fall short in accessing domain-specific or the latest knowledge. To mitigate this issue, retrieval-augmented generation (RAG) leveraging external knowledge bases (KBs), referred to as KB-VQA, emerges as a promising approach. Nevertheless, conventional unimodal retrieval techniques, which translate images into textual descriptions, oft"},"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":"2502.20964","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-02-28T11:25:38Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"5557e461b03afe5abc63612c13c3aa65c0376e237da826e3adb0fb7e7ccf04cd","abstract_canon_sha256":"93f9d67238d23937538561cfb267da3c486bc8f5f9981f6bc6760cc38ce5ef1d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:33:20.837518Z","signature_b64":"evbOcBKaJZhqsIC55KMZD2DIu5hLvxaH+cd6BD9lA1KMSL3aIyq//l57Izmhy9LbzG0rvHvUU/KA2pWUl+txDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8d3583df15ddaf3036e6030ccfbbea3511ed55c272e10ef5f91e1f90f56dc8c3","last_reissued_at":"2026-07-05T11:33:20.837024Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:33:20.837024Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Fine-Grained Knowledge Structuring and Retrieval for Visual Question Answering","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Nan Tang, Yin Wu, Yuyu Luo, Zhengxuan Zhang","submitted_at":"2025-02-28T11:25:38Z","abstract_excerpt":"Visual Question Answering (VQA) focuses on providing answers to natural language questions by utilizing information from images. Although cutting-edge multimodal large language models (MLLMs) such as GPT-4o achieve strong performance on VQA tasks, they frequently fall short in accessing domain-specific or the latest knowledge. To mitigate this issue, retrieval-augmented generation (RAG) leveraging external knowledge bases (KBs), referred to as KB-VQA, emerges as a promising approach. Nevertheless, conventional unimodal retrieval techniques, which translate images into textual descriptions, oft"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.20964","kind":"arxiv","version":3},"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/2502.20964/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":"2502.20964","created_at":"2026-07-05T11:33:20.837098+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.20964v3","created_at":"2026-07-05T11:33:20.837098+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.20964","created_at":"2026-07-05T11:33:20.837098+00:00"},{"alias_kind":"pith_short_12","alias_value":"RU2YHXYV3WXT","created_at":"2026-07-05T11:33:20.837098+00:00"},{"alias_kind":"pith_short_16","alias_value":"RU2YHXYV3WXTANXG","created_at":"2026-07-05T11:33:20.837098+00:00"},{"alias_kind":"pith_short_8","alias_value":"RU2YHXYV","created_at":"2026-07-05T11:33:20.837098+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.19307","citing_title":"MetaRA: Metamorphic Robustness Assessment for Multimodal Large Language Model-based Visual Question Answering Systems","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2605.03790","citing_title":"Enhancing Visual Question Answering with Multimodal LLMs via Chain-of-Question Guided Retrieval-Augmented Generation","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17889","citing_title":"AeroRAG: Structured Multimodal Retrieval-Augmented LLM for Fine-Grained Aerial Visual Reasoning","ref_index":24,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RU2YHXYV3WXTANXGAMGM7O7KGU","json":"https://pith.science/pith/RU2YHXYV3WXTANXGAMGM7O7KGU.json","graph_json":"https://pith.science/api/pith-number/RU2YHXYV3WXTANXGAMGM7O7KGU/graph.json","events_json":"https://pith.science/api/pith-number/RU2YHXYV3WXTANXGAMGM7O7KGU/events.json","paper":"https://pith.science/paper/RU2YHXYV"},"agent_actions":{"view_html":"https://pith.science/pith/RU2YHXYV3WXTANXGAMGM7O7KGU","download_json":"https://pith.science/pith/RU2YHXYV3WXTANXGAMGM7O7KGU.json","view_paper":"https://pith.science/paper/RU2YHXYV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.20964&json=true","fetch_graph":"https://pith.science/api/pith-number/RU2YHXYV3WXTANXGAMGM7O7KGU/graph.json","fetch_events":"https://pith.science/api/pith-number/RU2YHXYV3WXTANXGAMGM7O7KGU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RU2YHXYV3WXTANXGAMGM7O7KGU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RU2YHXYV3WXTANXGAMGM7O7KGU/action/storage_attestation","attest_author":"https://pith.science/pith/RU2YHXYV3WXTANXGAMGM7O7KGU/action/author_attestation","sign_citation":"https://pith.science/pith/RU2YHXYV3WXTANXGAMGM7O7KGU/action/citation_signature","submit_replication":"https://pith.science/pith/RU2YHXYV3WXTANXGAMGM7O7KGU/action/replication_record"}},"created_at":"2026-07-05T11:33:20.837098+00:00","updated_at":"2026-07-05T11:33:20.837098+00:00"}