{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:OSOJ7YRGWP7PX3L2T7LX2YU42S","short_pith_number":"pith:OSOJ7YRG","schema_version":"1.0","canonical_sha256":"749c9fe226b3fefbed7a9fd77d629cd4b4726b8122b88cb2b239be2427389ac0","source":{"kind":"arxiv","id":"2408.08673","version":2},"attestation_state":"computed","paper":{"title":"MAT-SED: A Masked Audio Transformer with Masked-Reconstruction Based Pre-training for Sound Event Detection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","eess.AS"],"primary_cat":"cs.SD","authors_text":"Haoyu Song, Ian McLoughlin, Kang Li, Pengfei Cai, Yan Song","submitted_at":"2024-08-16T11:33:16Z","abstract_excerpt":"Sound event detection (SED) methods that leverage a large pre-trained Transformer encoder network have shown promising performance in recent DCASE challenges. However, they still rely on an RNN-based context network to model temporal dependencies, largely due to the scarcity of labeled data. In this work, we propose a pure Transformer-based SED model with masked-reconstruction based pre-training, termed MAT-SED. Specifically, a Transformer with relative positional encoding is first designed as the context network, pre-trained by the masked-reconstruction task on all available target data in a "},"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":"2408.08673","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2024-08-16T11:33:16Z","cross_cats_sorted":["cs.AI","eess.AS"],"title_canon_sha256":"5f46214a4b7f4aa3d6ac19fdd213d7ccffb55fec6dddb1660cfa7bf0a8b0f4a1","abstract_canon_sha256":"2975704c73e0ae19d18318df1f143bfd296fe14027e096a5c5dfd993df161b45"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:56:24.824473Z","signature_b64":"u4Jk+Pg/UaaNFpd9/2Wm/9sDbqSjnNqRaHCYq59FrUsx1AgObhG8ofx38csIykkE7otV9pN2g2j9j6y2bEGLAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"749c9fe226b3fefbed7a9fd77d629cd4b4726b8122b88cb2b239be2427389ac0","last_reissued_at":"2026-07-05T08:56:24.824027Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:56:24.824027Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MAT-SED: A Masked Audio Transformer with Masked-Reconstruction Based Pre-training for Sound Event Detection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","eess.AS"],"primary_cat":"cs.SD","authors_text":"Haoyu Song, Ian McLoughlin, Kang Li, Pengfei Cai, Yan Song","submitted_at":"2024-08-16T11:33:16Z","abstract_excerpt":"Sound event detection (SED) methods that leverage a large pre-trained Transformer encoder network have shown promising performance in recent DCASE challenges. However, they still rely on an RNN-based context network to model temporal dependencies, largely due to the scarcity of labeled data. In this work, we propose a pure Transformer-based SED model with masked-reconstruction based pre-training, termed MAT-SED. Specifically, a Transformer with relative positional encoding is first designed as the context network, pre-trained by the masked-reconstruction task on all available target data in a "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.08673","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/2408.08673/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":"2408.08673","created_at":"2026-07-05T08:56:24.824084+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.08673v2","created_at":"2026-07-05T08:56:24.824084+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.08673","created_at":"2026-07-05T08:56:24.824084+00:00"},{"alias_kind":"pith_short_12","alias_value":"OSOJ7YRGWP7P","created_at":"2026-07-05T08:56:24.824084+00:00"},{"alias_kind":"pith_short_16","alias_value":"OSOJ7YRGWP7PX3L2","created_at":"2026-07-05T08:56:24.824084+00:00"},{"alias_kind":"pith_short_8","alias_value":"OSOJ7YRG","created_at":"2026-07-05T08:56:24.824084+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.00387","citing_title":"From Objectives to Applications: Aligning Architectural Biases in Audio Self-Supervised Learning","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OSOJ7YRGWP7PX3L2T7LX2YU42S","json":"https://pith.science/pith/OSOJ7YRGWP7PX3L2T7LX2YU42S.json","graph_json":"https://pith.science/api/pith-number/OSOJ7YRGWP7PX3L2T7LX2YU42S/graph.json","events_json":"https://pith.science/api/pith-number/OSOJ7YRGWP7PX3L2T7LX2YU42S/events.json","paper":"https://pith.science/paper/OSOJ7YRG"},"agent_actions":{"view_html":"https://pith.science/pith/OSOJ7YRGWP7PX3L2T7LX2YU42S","download_json":"https://pith.science/pith/OSOJ7YRGWP7PX3L2T7LX2YU42S.json","view_paper":"https://pith.science/paper/OSOJ7YRG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.08673&json=true","fetch_graph":"https://pith.science/api/pith-number/OSOJ7YRGWP7PX3L2T7LX2YU42S/graph.json","fetch_events":"https://pith.science/api/pith-number/OSOJ7YRGWP7PX3L2T7LX2YU42S/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OSOJ7YRGWP7PX3L2T7LX2YU42S/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OSOJ7YRGWP7PX3L2T7LX2YU42S/action/storage_attestation","attest_author":"https://pith.science/pith/OSOJ7YRGWP7PX3L2T7LX2YU42S/action/author_attestation","sign_citation":"https://pith.science/pith/OSOJ7YRGWP7PX3L2T7LX2YU42S/action/citation_signature","submit_replication":"https://pith.science/pith/OSOJ7YRGWP7PX3L2T7LX2YU42S/action/replication_record"}},"created_at":"2026-07-05T08:56:24.824084+00:00","updated_at":"2026-07-05T08:56:24.824084+00:00"}