{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:CBG6ML7TXIJELKW3FF6FR7W4JO","short_pith_number":"pith:CBG6ML7T","schema_version":"1.0","canonical_sha256":"104de62ff3ba1245aadb297c58fedc4bbe9fa3273a33c7aadb82baf3f6635509","source":{"kind":"arxiv","id":"2411.15913","version":4},"attestation_state":"computed","paper":{"title":"Repurposing Image Diffusion Models for Training-Free Music Style Transfer on Mel-spectrograms","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","eess.AS"],"primary_cat":"cs.SD","authors_text":"Heehwan Wang, Jiook Cha, Joonwoo Kwon, Jungwoo Seo, Shinjae Yoo, Sooyoung Kim, Yuewei Lin","submitted_at":"2024-11-24T16:53:34Z","abstract_excerpt":"Music style transfer blends source structure with reference style to enable personalized music creation. However, existing zero-shot methods often struggle to capture fine-grained audio nuances, relying on coarse text descriptions or requiring expensive task-specific training. We propose Stylus, a training-free framework that repurposes pretrained image diffusion models for music style transfer in the Mel-spectrogram domain. By treating audio as structured time-frequency images, Stylus manipulates self-attention by injecting style keys and values while preserving source structural queries. To "},"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":"2411.15913","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2024-11-24T16:53:34Z","cross_cats_sorted":["cs.AI","cs.LG","eess.AS"],"title_canon_sha256":"01b41f410aa48e17c27ef490a5e6258191a3a72990af6caf3e5fb947c9a06435","abstract_canon_sha256":"7fadb84514cb0aa3a2df639effb9d5beb196efc071800c4fafd67ff0d4200ea7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T02:45:14.710991Z","signature_b64":"QIQE6vbLB0czu9lGIDsPiZX9B/1lXl6FrI6z9mKuKYQUycqAG+CwQCtgCY6ngLKwE6qK9waFxj/Wac1CKBQtBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"104de62ff3ba1245aadb297c58fedc4bbe9fa3273a33c7aadb82baf3f6635509","last_reissued_at":"2026-05-18T02:45:14.710342Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T02:45:14.710342Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Repurposing Image Diffusion Models for Training-Free Music Style Transfer on Mel-spectrograms","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","eess.AS"],"primary_cat":"cs.SD","authors_text":"Heehwan Wang, Jiook Cha, Joonwoo Kwon, Jungwoo Seo, Shinjae Yoo, Sooyoung Kim, Yuewei Lin","submitted_at":"2024-11-24T16:53:34Z","abstract_excerpt":"Music style transfer blends source structure with reference style to enable personalized music creation. However, existing zero-shot methods often struggle to capture fine-grained audio nuances, relying on coarse text descriptions or requiring expensive task-specific training. We propose Stylus, a training-free framework that repurposes pretrained image diffusion models for music style transfer in the Mel-spectrogram domain. By treating audio as structured time-frequency images, Stylus manipulates self-attention by injecting style keys and values while preserving source structural queries. To "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.15913","kind":"arxiv","version":4},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"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":"2411.15913","created_at":"2026-05-18T02:45:14.710450+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.15913v4","created_at":"2026-05-18T02:45:14.710450+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.15913","created_at":"2026-05-18T02:45:14.710450+00:00"},{"alias_kind":"pith_short_12","alias_value":"CBG6ML7TXIJE","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"CBG6ML7TXIJELKW3","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"CBG6ML7T","created_at":"2026-05-18T12:33:37.589309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2411.15913","citing_title":"Repurposing Image Diffusion Models for Training-Free Music Style Transfer on Mel-spectrograms","ref_index":2,"is_internal_anchor":true},{"citing_arxiv_id":"2604.17986","citing_title":"Latent Fourier Transform","ref_index":40,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CBG6ML7TXIJELKW3FF6FR7W4JO","json":"https://pith.science/pith/CBG6ML7TXIJELKW3FF6FR7W4JO.json","graph_json":"https://pith.science/api/pith-number/CBG6ML7TXIJELKW3FF6FR7W4JO/graph.json","events_json":"https://pith.science/api/pith-number/CBG6ML7TXIJELKW3FF6FR7W4JO/events.json","paper":"https://pith.science/paper/CBG6ML7T"},"agent_actions":{"view_html":"https://pith.science/pith/CBG6ML7TXIJELKW3FF6FR7W4JO","download_json":"https://pith.science/pith/CBG6ML7TXIJELKW3FF6FR7W4JO.json","view_paper":"https://pith.science/paper/CBG6ML7T","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.15913&json=true","fetch_graph":"https://pith.science/api/pith-number/CBG6ML7TXIJELKW3FF6FR7W4JO/graph.json","fetch_events":"https://pith.science/api/pith-number/CBG6ML7TXIJELKW3FF6FR7W4JO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CBG6ML7TXIJELKW3FF6FR7W4JO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CBG6ML7TXIJELKW3FF6FR7W4JO/action/storage_attestation","attest_author":"https://pith.science/pith/CBG6ML7TXIJELKW3FF6FR7W4JO/action/author_attestation","sign_citation":"https://pith.science/pith/CBG6ML7TXIJELKW3FF6FR7W4JO/action/citation_signature","submit_replication":"https://pith.science/pith/CBG6ML7TXIJELKW3FF6FR7W4JO/action/replication_record"}},"created_at":"2026-05-18T02:45:14.710450+00:00","updated_at":"2026-05-18T02:45:14.710450+00:00"}