{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:73ZNKMVXYMQEUM3NTKEIRV4EMY","short_pith_number":"pith:73ZNKMVX","schema_version":"1.0","canonical_sha256":"fef2d532b7c3204a336d9a8888d7846618a867a6b7e40068656d5c7d56701653","source":{"kind":"arxiv","id":"2410.11522","version":2},"attestation_state":"computed","paper":{"title":"Leveraging LLM Embeddings for Cross Dataset Label Alignment and Zero Shot Music Emotion Prediction","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.LG","cs.MM","eess.AS"],"primary_cat":"cs.SD","authors_text":"Abhinaba Roy, Dorien Herremans, Renhang Liu","submitted_at":"2024-10-15T11:48:31Z","abstract_excerpt":"In this work, we present a novel method for music emotion recognition that leverages Large Language Model (LLM) embeddings for label alignment across multiple datasets and zero-shot prediction on novel categories. First, we compute LLM embeddings for emotion labels and apply non-parametric clustering to group similar labels, across multiple datasets containing disjoint labels. We use these cluster centers to map music features (MERT) to the LLM embedding space. To further enhance the model, we introduce an alignment regularization that enables dissociation of MERT embeddings from different clu"},"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":"2410.11522","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.SD","submitted_at":"2024-10-15T11:48:31Z","cross_cats_sorted":["cs.LG","cs.MM","eess.AS"],"title_canon_sha256":"62034aa25414ae966e0a9a6210c580416ba7ff66aba7ca28e35848b5932f38f8","abstract_canon_sha256":"e5c4114f384b1b6802653b76fb0896bc485d8b01e02ce58f9470d272a64f9222"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:21:50.181763Z","signature_b64":"kTbdM1oZM3J1QUfooWL6q0lsIYEPqRSrNLPJ/aWr6C8BnIQ14kSAk6WNweVwk80DWtwkprmLgIDd54uxCA/6Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fef2d532b7c3204a336d9a8888d7846618a867a6b7e40068656d5c7d56701653","last_reissued_at":"2026-07-05T09:21:50.181272Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:21:50.181272Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Leveraging LLM Embeddings for Cross Dataset Label Alignment and Zero Shot Music Emotion Prediction","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.LG","cs.MM","eess.AS"],"primary_cat":"cs.SD","authors_text":"Abhinaba Roy, Dorien Herremans, Renhang Liu","submitted_at":"2024-10-15T11:48:31Z","abstract_excerpt":"In this work, we present a novel method for music emotion recognition that leverages Large Language Model (LLM) embeddings for label alignment across multiple datasets and zero-shot prediction on novel categories. First, we compute LLM embeddings for emotion labels and apply non-parametric clustering to group similar labels, across multiple datasets containing disjoint labels. We use these cluster centers to map music features (MERT) to the LLM embedding space. To further enhance the model, we introduce an alignment regularization that enables dissociation of MERT embeddings from different clu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.11522","kind":"arxiv","version":2},"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/2410.11522/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":"2410.11522","created_at":"2026-07-05T09:21:50.181340+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.11522v2","created_at":"2026-07-05T09:21:50.181340+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.11522","created_at":"2026-07-05T09:21:50.181340+00:00"},{"alias_kind":"pith_short_12","alias_value":"73ZNKMVXYMQE","created_at":"2026-07-05T09:21:50.181340+00:00"},{"alias_kind":"pith_short_16","alias_value":"73ZNKMVXYMQEUM3N","created_at":"2026-07-05T09:21:50.181340+00:00"},{"alias_kind":"pith_short_8","alias_value":"73ZNKMVX","created_at":"2026-07-05T09:21:50.181340+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2606.21979","citing_title":"Toward Open-Set Speaker Attribute Prediction with Keyword-Appended LLM Embeddings","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2605.27346","citing_title":"MERIT: Learning Disentangled Music Representations for Audio Similarity","ref_index":27,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/73ZNKMVXYMQEUM3NTKEIRV4EMY","json":"https://pith.science/pith/73ZNKMVXYMQEUM3NTKEIRV4EMY.json","graph_json":"https://pith.science/api/pith-number/73ZNKMVXYMQEUM3NTKEIRV4EMY/graph.json","events_json":"https://pith.science/api/pith-number/73ZNKMVXYMQEUM3NTKEIRV4EMY/events.json","paper":"https://pith.science/paper/73ZNKMVX"},"agent_actions":{"view_html":"https://pith.science/pith/73ZNKMVXYMQEUM3NTKEIRV4EMY","download_json":"https://pith.science/pith/73ZNKMVXYMQEUM3NTKEIRV4EMY.json","view_paper":"https://pith.science/paper/73ZNKMVX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.11522&json=true","fetch_graph":"https://pith.science/api/pith-number/73ZNKMVXYMQEUM3NTKEIRV4EMY/graph.json","fetch_events":"https://pith.science/api/pith-number/73ZNKMVXYMQEUM3NTKEIRV4EMY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/73ZNKMVXYMQEUM3NTKEIRV4EMY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/73ZNKMVXYMQEUM3NTKEIRV4EMY/action/storage_attestation","attest_author":"https://pith.science/pith/73ZNKMVXYMQEUM3NTKEIRV4EMY/action/author_attestation","sign_citation":"https://pith.science/pith/73ZNKMVXYMQEUM3NTKEIRV4EMY/action/citation_signature","submit_replication":"https://pith.science/pith/73ZNKMVXYMQEUM3NTKEIRV4EMY/action/replication_record"}},"created_at":"2026-07-05T09:21:50.181340+00:00","updated_at":"2026-07-05T09:21:50.181340+00:00"}