{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:MBSTGMMVN4SZNHC3SPHTFEO7KN","short_pith_number":"pith:MBSTGMMV","schema_version":"1.0","canonical_sha256":"60653331956f25969c5b93cf3291df53582aa43904401f5873cd1b1869f83a38","source":{"kind":"arxiv","id":"2208.12415","version":1},"attestation_state":"computed","paper":{"title":"MuLan: A Joint Embedding of Music Audio and Natural Language","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.SD","stat.ML"],"primary_cat":"eess.AS","authors_text":"Aren Jansen, Daniel P. W. Ellis, Joonseok Lee, Judith Yue Li, Qingqing Huang, Ravi Ganti","submitted_at":"2022-08-26T03:13:21Z","abstract_excerpt":"Music tagging and content-based retrieval systems have traditionally been constructed using pre-defined ontologies covering a rigid set of music attributes or text queries. This paper presents MuLan: a first attempt at a new generation of acoustic models that link music audio directly to unconstrained natural language music descriptions. MuLan takes the form of a two-tower, joint audio-text embedding model trained using 44 million music recordings (370K hours) and weakly-associated, free-form text annotations. Through its compatibility with a wide range of music genres and text styles (includi"},"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":"2208.12415","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2022-08-26T03:13:21Z","cross_cats_sorted":["cs.CL","cs.SD","stat.ML"],"title_canon_sha256":"12118640b1cae47ad5510e0543a5c65cd48d49e7f35880e8fa776f8ce27c797e","abstract_canon_sha256":"aecaa4c1cf43a698a706d79bea589eeaac5d6460209098687561111efeb4fa62"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:51:45.023802Z","signature_b64":"I9wbx8az2F7e63CwwunL5rSew2EejQ1Yl77uFNhknaN1n/xY7WU6UW5v0418p0kewQX3YrE+UbW8i9O5SF/iCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"60653331956f25969c5b93cf3291df53582aa43904401f5873cd1b1869f83a38","last_reissued_at":"2026-07-05T04:51:45.023380Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:51:45.023380Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MuLan: A Joint Embedding of Music Audio and Natural Language","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.SD","stat.ML"],"primary_cat":"eess.AS","authors_text":"Aren Jansen, Daniel P. W. Ellis, Joonseok Lee, Judith Yue Li, Qingqing Huang, Ravi Ganti","submitted_at":"2022-08-26T03:13:21Z","abstract_excerpt":"Music tagging and content-based retrieval systems have traditionally been constructed using pre-defined ontologies covering a rigid set of music attributes or text queries. This paper presents MuLan: a first attempt at a new generation of acoustic models that link music audio directly to unconstrained natural language music descriptions. MuLan takes the form of a two-tower, joint audio-text embedding model trained using 44 million music recordings (370K hours) and weakly-associated, free-form text annotations. Through its compatibility with a wide range of music genres and text styles (includi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2208.12415","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/2208.12415/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":"2208.12415","created_at":"2026-07-05T04:51:45.023438+00:00"},{"alias_kind":"arxiv_version","alias_value":"2208.12415v1","created_at":"2026-07-05T04:51:45.023438+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2208.12415","created_at":"2026-07-05T04:51:45.023438+00:00"},{"alias_kind":"pith_short_12","alias_value":"MBSTGMMVN4SZ","created_at":"2026-07-05T04:51:45.023438+00:00"},{"alias_kind":"pith_short_16","alias_value":"MBSTGMMVN4SZNHC3","created_at":"2026-07-05T04:51:45.023438+00:00"},{"alias_kind":"pith_short_8","alias_value":"MBSTGMMV","created_at":"2026-07-05T04:51:45.023438+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.06615","citing_title":"FIGMA: Towards FIne-Grained Music retrievAl","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2606.03169","citing_title":"SketchSong: Hierarchical Song Generation with Sketch Planning and Fine-Grained Multi-Track Modeling","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06582","citing_title":"PairAlign: A Framework for Sequence Tokenization via Self-Alignment with Applications to Audio Tokenization","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2605.27346","citing_title":"MERIT: Learning Disentangled Music Representations for Audio Similarity","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2603.03190","citing_title":"Expectation and Acoustic Neural Network Representations Enhance Music Identification from Brain Activity","ref_index":62,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21081","citing_title":"Musical Attention Transformer: Music Generation Using a Music-Specific Attention Model","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06582","citing_title":"PairAlign: A Framework for Sequence Tokenization via Self-Alignment with Applications to Audio Tokenization","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00824","citing_title":"CustomDancer: Customized Dance Recommendation by Text-Dance Retrieval","ref_index":37,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MBSTGMMVN4SZNHC3SPHTFEO7KN","json":"https://pith.science/pith/MBSTGMMVN4SZNHC3SPHTFEO7KN.json","graph_json":"https://pith.science/api/pith-number/MBSTGMMVN4SZNHC3SPHTFEO7KN/graph.json","events_json":"https://pith.science/api/pith-number/MBSTGMMVN4SZNHC3SPHTFEO7KN/events.json","paper":"https://pith.science/paper/MBSTGMMV"},"agent_actions":{"view_html":"https://pith.science/pith/MBSTGMMVN4SZNHC3SPHTFEO7KN","download_json":"https://pith.science/pith/MBSTGMMVN4SZNHC3SPHTFEO7KN.json","view_paper":"https://pith.science/paper/MBSTGMMV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2208.12415&json=true","fetch_graph":"https://pith.science/api/pith-number/MBSTGMMVN4SZNHC3SPHTFEO7KN/graph.json","fetch_events":"https://pith.science/api/pith-number/MBSTGMMVN4SZNHC3SPHTFEO7KN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MBSTGMMVN4SZNHC3SPHTFEO7KN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MBSTGMMVN4SZNHC3SPHTFEO7KN/action/storage_attestation","attest_author":"https://pith.science/pith/MBSTGMMVN4SZNHC3SPHTFEO7KN/action/author_attestation","sign_citation":"https://pith.science/pith/MBSTGMMVN4SZNHC3SPHTFEO7KN/action/citation_signature","submit_replication":"https://pith.science/pith/MBSTGMMVN4SZNHC3SPHTFEO7KN/action/replication_record"}},"created_at":"2026-07-05T04:51:45.023438+00:00","updated_at":"2026-07-05T04:51:45.023438+00:00"}