{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:FLDF2O7YLRLTKXIIGOMUWSX7KJ","short_pith_number":"pith:FLDF2O7Y","schema_version":"1.0","canonical_sha256":"2ac65d3bf85c57355d0833994b4aff526ab55483e14dead81e697ea5c218142b","source":{"kind":"arxiv","id":"2503.24193","version":2},"attestation_state":"computed","paper":{"title":"Text2Tracks: Prompt-based Music Recommendation via Generative Retrieval","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Alice Wang, Andreas Damianou, Enrico Palumbo, Gustavo Penha, Hugues Bouchard, Jos\\'e Luis Redondo Garc\\'ia, Mounia Lalmas, Timothy Christopher Heath","submitted_at":"2025-03-31T15:09:19Z","abstract_excerpt":"In recent years, Large Language Models (LLMs) have enabled users to provide highly specific music recommendation requests using natural language prompts (e.g. \"Can you recommend some old classics for slow dancing?\"). In this setup, the recommended tracks are predicted by the LLM in an autoregressive way, i.e. the LLM generates the track titles one token at a time. While intuitive, this approach has several limitation. First, it is based on a general purpose tokenization that is optimized for words rather than for track titles. Second, it necessitates an additional entity resolution layer that "},"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":"2503.24193","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.IR","submitted_at":"2025-03-31T15:09:19Z","cross_cats_sorted":[],"title_canon_sha256":"d48ec6659f17d5421d417cf3db97b508f359402715d22a8d87af34039447aa8e","abstract_canon_sha256":"0f224071305c9f65cc0726e72eb90a9662a0bd4e6060e9aff999e4f160db9b4d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:43:27.279017Z","signature_b64":"Huazjd56lt5SWQDooI4nWVEXj0MVtW0PzeJnKNljVcRHjBdV7GEqAvlEDafcGLygVV8z0PxVWVhngbnT590TDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2ac65d3bf85c57355d0833994b4aff526ab55483e14dead81e697ea5c218142b","last_reissued_at":"2026-07-05T10:43:27.278383Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:43:27.278383Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Text2Tracks: Prompt-based Music Recommendation via Generative Retrieval","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Alice Wang, Andreas Damianou, Enrico Palumbo, Gustavo Penha, Hugues Bouchard, Jos\\'e Luis Redondo Garc\\'ia, Mounia Lalmas, Timothy Christopher Heath","submitted_at":"2025-03-31T15:09:19Z","abstract_excerpt":"In recent years, Large Language Models (LLMs) have enabled users to provide highly specific music recommendation requests using natural language prompts (e.g. \"Can you recommend some old classics for slow dancing?\"). In this setup, the recommended tracks are predicted by the LLM in an autoregressive way, i.e. the LLM generates the track titles one token at a time. While intuitive, this approach has several limitation. First, it is based on a general purpose tokenization that is optimized for words rather than for track titles. Second, it necessitates an additional entity resolution layer that "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.24193","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/2503.24193/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":"2503.24193","created_at":"2026-07-05T10:43:27.278469+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.24193v2","created_at":"2026-07-05T10:43:27.278469+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.24193","created_at":"2026-07-05T10:43:27.278469+00:00"},{"alias_kind":"pith_short_12","alias_value":"FLDF2O7YLRLT","created_at":"2026-07-05T10:43:27.278469+00:00"},{"alias_kind":"pith_short_16","alias_value":"FLDF2O7YLRLTKXII","created_at":"2026-07-05T10:43:27.278469+00:00"},{"alias_kind":"pith_short_8","alias_value":"FLDF2O7Y","created_at":"2026-07-05T10:43:27.278469+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.09120","citing_title":"Reddit2Deezer: A Scalable Dataset for Real-World Grounded Conversational Music Recommendation","ref_index":24,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FLDF2O7YLRLTKXIIGOMUWSX7KJ","json":"https://pith.science/pith/FLDF2O7YLRLTKXIIGOMUWSX7KJ.json","graph_json":"https://pith.science/api/pith-number/FLDF2O7YLRLTKXIIGOMUWSX7KJ/graph.json","events_json":"https://pith.science/api/pith-number/FLDF2O7YLRLTKXIIGOMUWSX7KJ/events.json","paper":"https://pith.science/paper/FLDF2O7Y"},"agent_actions":{"view_html":"https://pith.science/pith/FLDF2O7YLRLTKXIIGOMUWSX7KJ","download_json":"https://pith.science/pith/FLDF2O7YLRLTKXIIGOMUWSX7KJ.json","view_paper":"https://pith.science/paper/FLDF2O7Y","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.24193&json=true","fetch_graph":"https://pith.science/api/pith-number/FLDF2O7YLRLTKXIIGOMUWSX7KJ/graph.json","fetch_events":"https://pith.science/api/pith-number/FLDF2O7YLRLTKXIIGOMUWSX7KJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FLDF2O7YLRLTKXIIGOMUWSX7KJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FLDF2O7YLRLTKXIIGOMUWSX7KJ/action/storage_attestation","attest_author":"https://pith.science/pith/FLDF2O7YLRLTKXIIGOMUWSX7KJ/action/author_attestation","sign_citation":"https://pith.science/pith/FLDF2O7YLRLTKXIIGOMUWSX7KJ/action/citation_signature","submit_replication":"https://pith.science/pith/FLDF2O7YLRLTKXIIGOMUWSX7KJ/action/replication_record"}},"created_at":"2026-07-05T10:43:27.278469+00:00","updated_at":"2026-07-05T10:43:27.278469+00:00"}