{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:HN3GWO7XATOIA36BHDWSAVDZXJ","short_pith_number":"pith:HN3GWO7X","schema_version":"1.0","canonical_sha256":"3b766b3bf704dc806fc138ed205479ba7e5b053fb469b4612ae46c7d6d59a249","source":{"kind":"arxiv","id":"2605.16403","version":1},"attestation_state":"computed","paper":{"title":"When Vision Speaks for Sound","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"cs.CV","authors_text":"Muhao Chen, Peng Qi, Rui Cai, Tinghui Zhu, Wendi Li, Wenjie Jacky Mo, Xiaofei Wen, Xingyu Fu, Yanan Xie","submitted_at":"2026-05-13T05:00:19Z","abstract_excerpt":"Despite rapid progress in video-capable MLLMs, we find that their apparent audio understanding in videos is often vision-driven: models rely on visual cues to infer or hallucinate acoustic information, rather than verifying the audio stream. This issue appears across both state-of-the-art open-source omni models and leading closed-source models from providers such as Google and OpenAI. We characterize this failure mode as an audio-visual Clever Hans effect, in which models appear (falsely) audio-grounded, but actually exploit visual-acoustic correlations without verifying whether the audio and"},"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":"2605.16403","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-13T05:00:19Z","cross_cats_sorted":["cs.SD"],"title_canon_sha256":"52acd4cd7a1b8f90645672ba808d208d4cafa2473b1e943468196073837e18b4","abstract_canon_sha256":"1fd692a43af4dbfe265c9f71763f446d16ca4de7cc5a489ad4b044c6d1037f02"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:02:20.586169Z","signature_b64":"y5UwVa7Q4Ctnv/hsnkKBHLV9tNP7dAtCWWGqE+qNUjzn0BMdoti+M60OsmMXeZHLvLxkHVuEFXyen/T+9mnWDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3b766b3bf704dc806fc138ed205479ba7e5b053fb469b4612ae46c7d6d59a249","last_reissued_at":"2026-05-20T00:02:20.585408Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:02:20.585408Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"When Vision Speaks for Sound","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD"],"primary_cat":"cs.CV","authors_text":"Muhao Chen, Peng Qi, Rui Cai, Tinghui Zhu, Wendi Li, Wenjie Jacky Mo, Xiaofei Wen, Xingyu Fu, Yanan Xie","submitted_at":"2026-05-13T05:00:19Z","abstract_excerpt":"Despite rapid progress in video-capable MLLMs, we find that their apparent audio understanding in videos is often vision-driven: models rely on visual cues to infer or hallucinate acoustic information, rather than verifying the audio stream. This issue appears across both state-of-the-art open-source omni models and leading closed-source models from providers such as Google and OpenAI. We characterize this failure mode as an audio-visual Clever Hans effect, in which models appear (falsely) audio-grounded, but actually exploit visual-acoustic correlations without verifying whether the audio and"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.16403","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/2605.16403/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-19T20:21:57.657330Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T19:34:36.601172Z","status":"skipped","version":"1.0.0","findings_count":0}],"snapshot_sha256":"8dee41cc6b0bc83d045b13995bb79afab260e5582bc1bebc4978ac5f3c12001b"},"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":"2605.16403","created_at":"2026-05-20T00:02:20.585528+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.16403v1","created_at":"2026-05-20T00:02:20.585528+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.16403","created_at":"2026-05-20T00:02:20.585528+00:00"},{"alias_kind":"pith_short_12","alias_value":"HN3GWO7XATOI","created_at":"2026-05-20T00:02:20.585528+00:00"},{"alias_kind":"pith_short_16","alias_value":"HN3GWO7XATOIA36B","created_at":"2026-05-20T00:02:20.585528+00:00"},{"alias_kind":"pith_short_8","alias_value":"HN3GWO7X","created_at":"2026-05-20T00:02:20.585528+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/HN3GWO7XATOIA36BHDWSAVDZXJ","json":"https://pith.science/pith/HN3GWO7XATOIA36BHDWSAVDZXJ.json","graph_json":"https://pith.science/api/pith-number/HN3GWO7XATOIA36BHDWSAVDZXJ/graph.json","events_json":"https://pith.science/api/pith-number/HN3GWO7XATOIA36BHDWSAVDZXJ/events.json","paper":"https://pith.science/paper/HN3GWO7X"},"agent_actions":{"view_html":"https://pith.science/pith/HN3GWO7XATOIA36BHDWSAVDZXJ","download_json":"https://pith.science/pith/HN3GWO7XATOIA36BHDWSAVDZXJ.json","view_paper":"https://pith.science/paper/HN3GWO7X","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.16403&json=true","fetch_graph":"https://pith.science/api/pith-number/HN3GWO7XATOIA36BHDWSAVDZXJ/graph.json","fetch_events":"https://pith.science/api/pith-number/HN3GWO7XATOIA36BHDWSAVDZXJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HN3GWO7XATOIA36BHDWSAVDZXJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HN3GWO7XATOIA36BHDWSAVDZXJ/action/storage_attestation","attest_author":"https://pith.science/pith/HN3GWO7XATOIA36BHDWSAVDZXJ/action/author_attestation","sign_citation":"https://pith.science/pith/HN3GWO7XATOIA36BHDWSAVDZXJ/action/citation_signature","submit_replication":"https://pith.science/pith/HN3GWO7XATOIA36BHDWSAVDZXJ/action/replication_record"}},"created_at":"2026-05-20T00:02:20.585528+00:00","updated_at":"2026-05-20T00:02:20.585528+00:00"}