{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:U64TJUDLUSU3ADBVKK5NBCM6WO","short_pith_number":"pith:U64TJUDL","schema_version":"1.0","canonical_sha256":"a7b934d06ba4a9b00c3552bad0899eb3931c7a60f2f76fc7bc085625b12633d7","source":{"kind":"arxiv","id":"2505.06803","version":1},"attestation_state":"computed","paper":{"title":"Bridging Ears and Eyes: Analyzing Audio and Visual Large Language Models to Humans in Visible Sound Recognition and Reducing Their Sensory Gap via Cross-Modal Distillation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.CV","cs.MM","eess.AS"],"primary_cat":"cs.SD","authors_text":"Junkai Wu, Nima Mesgarani, Vishal Choudhari, Xilin Jiang","submitted_at":"2025-05-11T01:01:44Z","abstract_excerpt":"Audio large language models (LLMs) are considered experts at recognizing sound objects, yet their performance relative to LLMs in other sensory modalities, such as visual or audio-visual LLMs, and to humans using their ears, eyes, or both remains unexplored. To investigate this, we systematically evaluate audio, visual, and audio-visual LLMs, specifically Qwen2-Audio, Qwen2-VL, and Qwen2.5-Omni, against humans in recognizing sound objects of different classes from audio-only, silent video, or sounded video inputs. We uncover a performance gap between Qwen2-Audio and Qwen2-VL that parallels the"},"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":"2505.06803","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2025-05-11T01:01:44Z","cross_cats_sorted":["cs.CL","cs.CV","cs.MM","eess.AS"],"title_canon_sha256":"7c96d7c6760ddc6bff0aecef341c32514dbba61debfcd09493a5691b9a367bbc","abstract_canon_sha256":"1fd9151593173bbf1fa725b6c566a4a9b90f9098fa23ef1f3db8cf0467fe8100"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:01:15.990832Z","signature_b64":"jAygtKV9lREyoVt7SBhUPdB5ZD91leYB4kDLsxIC6Z50mvApWLF2Lo33V4mtl0WmkAy8Ks+TjD7bE4HfEAyeBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a7b934d06ba4a9b00c3552bad0899eb3931c7a60f2f76fc7bc085625b12633d7","last_reissued_at":"2026-07-05T11:01:15.990284Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:01:15.990284Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Bridging Ears and Eyes: Analyzing Audio and Visual Large Language Models to Humans in Visible Sound Recognition and Reducing Their Sensory Gap via Cross-Modal Distillation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.CV","cs.MM","eess.AS"],"primary_cat":"cs.SD","authors_text":"Junkai Wu, Nima Mesgarani, Vishal Choudhari, Xilin Jiang","submitted_at":"2025-05-11T01:01:44Z","abstract_excerpt":"Audio large language models (LLMs) are considered experts at recognizing sound objects, yet their performance relative to LLMs in other sensory modalities, such as visual or audio-visual LLMs, and to humans using their ears, eyes, or both remains unexplored. To investigate this, we systematically evaluate audio, visual, and audio-visual LLMs, specifically Qwen2-Audio, Qwen2-VL, and Qwen2.5-Omni, against humans in recognizing sound objects of different classes from audio-only, silent video, or sounded video inputs. We uncover a performance gap between Qwen2-Audio and Qwen2-VL that parallels the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.06803","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/2505.06803/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":"2505.06803","created_at":"2026-07-05T11:01:15.990362+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.06803v1","created_at":"2026-07-05T11:01:15.990362+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.06803","created_at":"2026-07-05T11:01:15.990362+00:00"},{"alias_kind":"pith_short_12","alias_value":"U64TJUDLUSU3","created_at":"2026-07-05T11:01:15.990362+00:00"},{"alias_kind":"pith_short_16","alias_value":"U64TJUDLUSU3ADBV","created_at":"2026-07-05T11:01:15.990362+00:00"},{"alias_kind":"pith_short_8","alias_value":"U64TJUDL","created_at":"2026-07-05T11:01:15.990362+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.10815","citing_title":"Probing Cross-modal Information Hubs in Audio-Visual LLMs","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10815","citing_title":"Probing Cross-modal Information Hubs in Audio-Visual LLMs","ref_index":3,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/U64TJUDLUSU3ADBVKK5NBCM6WO","json":"https://pith.science/pith/U64TJUDLUSU3ADBVKK5NBCM6WO.json","graph_json":"https://pith.science/api/pith-number/U64TJUDLUSU3ADBVKK5NBCM6WO/graph.json","events_json":"https://pith.science/api/pith-number/U64TJUDLUSU3ADBVKK5NBCM6WO/events.json","paper":"https://pith.science/paper/U64TJUDL"},"agent_actions":{"view_html":"https://pith.science/pith/U64TJUDLUSU3ADBVKK5NBCM6WO","download_json":"https://pith.science/pith/U64TJUDLUSU3ADBVKK5NBCM6WO.json","view_paper":"https://pith.science/paper/U64TJUDL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.06803&json=true","fetch_graph":"https://pith.science/api/pith-number/U64TJUDLUSU3ADBVKK5NBCM6WO/graph.json","fetch_events":"https://pith.science/api/pith-number/U64TJUDLUSU3ADBVKK5NBCM6WO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/U64TJUDLUSU3ADBVKK5NBCM6WO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/U64TJUDLUSU3ADBVKK5NBCM6WO/action/storage_attestation","attest_author":"https://pith.science/pith/U64TJUDLUSU3ADBVKK5NBCM6WO/action/author_attestation","sign_citation":"https://pith.science/pith/U64TJUDLUSU3ADBVKK5NBCM6WO/action/citation_signature","submit_replication":"https://pith.science/pith/U64TJUDLUSU3ADBVKK5NBCM6WO/action/replication_record"}},"created_at":"2026-07-05T11:01:15.990362+00:00","updated_at":"2026-07-05T11:01:15.990362+00:00"}