{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:K6F35A2S5WLUPKZGIN4XZSJWDL","short_pith_number":"pith:K6F35A2S","schema_version":"1.0","canonical_sha256":"578bbe8352ed9747ab2643797cc9361aefbb2a1ff0b8eedc4ed7abf2031a925b","source":{"kind":"arxiv","id":"2505.15559","version":1},"attestation_state":"computed","paper":{"title":"Moonbeam: A MIDI Foundation Model Using Both Absolute and Relative Music Attributes","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","eess.AS"],"primary_cat":"cs.SD","authors_text":"Simon Dixon, Zixun Guo","submitted_at":"2025-05-21T14:17:25Z","abstract_excerpt":"Moonbeam is a transformer-based foundation model for symbolic music, pretrained on a large and diverse collection of MIDI data totaling 81.6K hours of music and 18 billion tokens. Moonbeam incorporates music-domain inductive biases by capturing both absolute and relative musical attributes through the introduction of a novel domain-knowledge-inspired tokenization method and Multidimensional Relative Attention (MRA), which captures relative music information without additional trainable parameters. Leveraging the pretrained Moonbeam, we propose 2 finetuning architectures with full anticipatory "},"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":"2505.15559","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.SD","submitted_at":"2025-05-21T14:17:25Z","cross_cats_sorted":["cs.AI","eess.AS"],"title_canon_sha256":"de5e06343fed478128cb2eb2a4f4227b894258e2e4a9b64ae8e928f535cfc7ee","abstract_canon_sha256":"eec04c27d5c95103595dec3983d381a6ab7e1dae487455356eb8c99e776993e2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:06:47.679486Z","signature_b64":"0JEpgxYFoPBAlt2mXeVVQx2aEpoCO7ra0b8O4rFcywzfKmUFYvb7uR9Jh7yPpnCjhk8UNHy3xvOkrOVf2cymAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"578bbe8352ed9747ab2643797cc9361aefbb2a1ff0b8eedc4ed7abf2031a925b","last_reissued_at":"2026-07-05T11:06:47.678994Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:06:47.678994Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Moonbeam: A MIDI Foundation Model Using Both Absolute and Relative Music Attributes","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","eess.AS"],"primary_cat":"cs.SD","authors_text":"Simon Dixon, Zixun Guo","submitted_at":"2025-05-21T14:17:25Z","abstract_excerpt":"Moonbeam is a transformer-based foundation model for symbolic music, pretrained on a large and diverse collection of MIDI data totaling 81.6K hours of music and 18 billion tokens. Moonbeam incorporates music-domain inductive biases by capturing both absolute and relative musical attributes through the introduction of a novel domain-knowledge-inspired tokenization method and Multidimensional Relative Attention (MRA), which captures relative music information without additional trainable parameters. Leveraging the pretrained Moonbeam, we propose 2 finetuning architectures with full anticipatory "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.15559","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/2505.15559/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":"2505.15559","created_at":"2026-07-05T11:06:47.679054+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.15559v1","created_at":"2026-07-05T11:06:47.679054+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.15559","created_at":"2026-07-05T11:06:47.679054+00:00"},{"alias_kind":"pith_short_12","alias_value":"K6F35A2S5WLU","created_at":"2026-07-05T11:06:47.679054+00:00"},{"alias_kind":"pith_short_16","alias_value":"K6F35A2S5WLUPKZG","created_at":"2026-07-05T11:06:47.679054+00:00"},{"alias_kind":"pith_short_8","alias_value":"K6F35A2S","created_at":"2026-07-05T11:06:47.679054+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.19532","citing_title":"BEAT: Tokenizing and Generating Symbolic Music by Uniform Temporal Steps","ref_index":76,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/K6F35A2S5WLUPKZGIN4XZSJWDL","json":"https://pith.science/pith/K6F35A2S5WLUPKZGIN4XZSJWDL.json","graph_json":"https://pith.science/api/pith-number/K6F35A2S5WLUPKZGIN4XZSJWDL/graph.json","events_json":"https://pith.science/api/pith-number/K6F35A2S5WLUPKZGIN4XZSJWDL/events.json","paper":"https://pith.science/paper/K6F35A2S"},"agent_actions":{"view_html":"https://pith.science/pith/K6F35A2S5WLUPKZGIN4XZSJWDL","download_json":"https://pith.science/pith/K6F35A2S5WLUPKZGIN4XZSJWDL.json","view_paper":"https://pith.science/paper/K6F35A2S","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.15559&json=true","fetch_graph":"https://pith.science/api/pith-number/K6F35A2S5WLUPKZGIN4XZSJWDL/graph.json","fetch_events":"https://pith.science/api/pith-number/K6F35A2S5WLUPKZGIN4XZSJWDL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/K6F35A2S5WLUPKZGIN4XZSJWDL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/K6F35A2S5WLUPKZGIN4XZSJWDL/action/storage_attestation","attest_author":"https://pith.science/pith/K6F35A2S5WLUPKZGIN4XZSJWDL/action/author_attestation","sign_citation":"https://pith.science/pith/K6F35A2S5WLUPKZGIN4XZSJWDL/action/citation_signature","submit_replication":"https://pith.science/pith/K6F35A2S5WLUPKZGIN4XZSJWDL/action/replication_record"}},"created_at":"2026-07-05T11:06:47.679054+00:00","updated_at":"2026-07-05T11:06:47.679054+00:00"}