{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:23CI3LEOVWIAVPIAN2OU5BG67T","short_pith_number":"pith:23CI3LEO","schema_version":"1.0","canonical_sha256":"d6c48dac8ead900abd006e9d4e84defcd81b302d8067c42ee82c6295a14c3050","source":{"kind":"arxiv","id":"2605.30698","version":1},"attestation_state":"computed","paper":{"title":"Seeing Before Agreeing: Aligning Multi-Agent Consensus with Visual Evidence","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.MA"],"primary_cat":"cs.CV","authors_text":"Dongsheng Ma, Shaoxu Sun, Shuochen Chang, Wentao Zhang, Yalin Feng, Yikang Wang, Yinglong Yang, Yuanzi Li, Yufei Chen, Yuhan Wang, Zhengren Wang","submitted_at":"2026-05-29T00:45:25Z","abstract_excerpt":"Vision-language models (VLMs) have achieved strong performance on visual question answering (VQA). To mitigate individual hallucinations and blind spots, aggregating diverse perspectives via multi-agent collaboration has emerged as a promising paradigm. While this approach has shown great success in textual QA, its potential in the multimodal domain remains under-explored. Existing multi-agent VQA methods predominantly adapt text-centric protocols, focusing on textual discussions while ignoring the alignment of visual information. In this work, we reveal a key insight: answer-level agreement i"},"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":"2605.30698","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-29T00:45:25Z","cross_cats_sorted":["cs.AI","cs.MA"],"title_canon_sha256":"d7686646e4b04db14d1cf5f017ad90e2b8bafea701d7627639cc6fa8b65bb28c","abstract_canon_sha256":"1de58fb383e6c75c187e54e14ca903815ad05507e6bf991b12a8eea73d637f5d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-01T01:03:09.115001Z","signature_b64":"HBi8gL6v090ksWC0H8sPyLz8FP+KjDMsMwTIgLVaEsTPCLw37wLrWcngTcxlb2EfFgURgiAunoNBNekL1N69BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d6c48dac8ead900abd006e9d4e84defcd81b302d8067c42ee82c6295a14c3050","last_reissued_at":"2026-06-01T01:03:09.114332Z","signature_status":"signed_v1","first_computed_at":"2026-06-01T01:03:09.114332Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Seeing Before Agreeing: Aligning Multi-Agent Consensus with Visual Evidence","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.MA"],"primary_cat":"cs.CV","authors_text":"Dongsheng Ma, Shaoxu Sun, Shuochen Chang, Wentao Zhang, Yalin Feng, Yikang Wang, Yinglong Yang, Yuanzi Li, Yufei Chen, Yuhan Wang, Zhengren Wang","submitted_at":"2026-05-29T00:45:25Z","abstract_excerpt":"Vision-language models (VLMs) have achieved strong performance on visual question answering (VQA). To mitigate individual hallucinations and blind spots, aggregating diverse perspectives via multi-agent collaboration has emerged as a promising paradigm. While this approach has shown great success in textual QA, its potential in the multimodal domain remains under-explored. Existing multi-agent VQA methods predominantly adapt text-centric protocols, focusing on textual discussions while ignoring the alignment of visual information. In this work, we reveal a key insight: answer-level agreement i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.30698","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/2605.30698/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":"2605.30698","created_at":"2026-06-01T01:03:09.114458+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.30698v1","created_at":"2026-06-01T01:03:09.114458+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.30698","created_at":"2026-06-01T01:03:09.114458+00:00"},{"alias_kind":"pith_short_12","alias_value":"23CI3LEOVWIA","created_at":"2026-06-01T01:03:09.114458+00:00"},{"alias_kind":"pith_short_16","alias_value":"23CI3LEOVWIAVPIA","created_at":"2026-06-01T01:03:09.114458+00:00"},{"alias_kind":"pith_short_8","alias_value":"23CI3LEO","created_at":"2026-06-01T01:03:09.114458+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/23CI3LEOVWIAVPIAN2OU5BG67T","json":"https://pith.science/pith/23CI3LEOVWIAVPIAN2OU5BG67T.json","graph_json":"https://pith.science/api/pith-number/23CI3LEOVWIAVPIAN2OU5BG67T/graph.json","events_json":"https://pith.science/api/pith-number/23CI3LEOVWIAVPIAN2OU5BG67T/events.json","paper":"https://pith.science/paper/23CI3LEO"},"agent_actions":{"view_html":"https://pith.science/pith/23CI3LEOVWIAVPIAN2OU5BG67T","download_json":"https://pith.science/pith/23CI3LEOVWIAVPIAN2OU5BG67T.json","view_paper":"https://pith.science/paper/23CI3LEO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.30698&json=true","fetch_graph":"https://pith.science/api/pith-number/23CI3LEOVWIAVPIAN2OU5BG67T/graph.json","fetch_events":"https://pith.science/api/pith-number/23CI3LEOVWIAVPIAN2OU5BG67T/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/23CI3LEOVWIAVPIAN2OU5BG67T/action/timestamp_anchor","attest_storage":"https://pith.science/pith/23CI3LEOVWIAVPIAN2OU5BG67T/action/storage_attestation","attest_author":"https://pith.science/pith/23CI3LEOVWIAVPIAN2OU5BG67T/action/author_attestation","sign_citation":"https://pith.science/pith/23CI3LEOVWIAVPIAN2OU5BG67T/action/citation_signature","submit_replication":"https://pith.science/pith/23CI3LEOVWIAVPIAN2OU5BG67T/action/replication_record"}},"created_at":"2026-06-01T01:03:09.114458+00:00","updated_at":"2026-06-01T01:03:09.114458+00:00"}