{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:PY4GW77WHJS4TFYIW4EVYS7WD5","short_pith_number":"pith:PY4GW77W","schema_version":"1.0","canonical_sha256":"7e386b7ff63a65c99708b7095c4bf61f72b3ed9e9739806a7677b6eecd351e14","source":{"kind":"arxiv","id":"2207.06405","version":3},"attestation_state":"computed","paper":{"title":"Masked Autoencoders that Listen","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","eess.AS"],"primary_cat":"cs.SD","authors_text":"Alexei Baevski, Christoph Feichtenhofer, Florian Metze, Hu Xu, Juncheng Li, Michael Auli, Po-Yao Huang, Wojciech Galuba","submitted_at":"2022-07-13T17:59:55Z","abstract_excerpt":"This paper studies a simple extension of image-based Masked Autoencoders (MAE) to self-supervised representation learning from audio spectrograms. Following the Transformer encoder-decoder design in MAE, our Audio-MAE first encodes audio spectrogram patches with a high masking ratio, feeding only the non-masked tokens through encoder layers. The decoder then re-orders and decodes the encoded context padded with mask tokens, in order to reconstruct the input spectrogram. We find it beneficial to incorporate local window attention in the decoder, as audio spectrograms are highly correlated in lo"},"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":"2207.06405","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2022-07-13T17:59:55Z","cross_cats_sorted":["cs.AI","cs.LG","eess.AS"],"title_canon_sha256":"3af8f0d30ec36869ef9501cbc54b560916f866ded8d1554f7661b87ad9cebd6a","abstract_canon_sha256":"7ee002bca1e15d7b5b6de1c179b9a51c64ce49ee6117b7408b4e421d55fbef06"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:32:31.325438Z","signature_b64":"z1eCCoBL5r02iPaEsotBSdDldvrFAmliUiEOatInkggXIsYMc6k2+HXwkPMgJKbnn0LXSL30NjDzH9ZhVryJDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7e386b7ff63a65c99708b7095c4bf61f72b3ed9e9739806a7677b6eecd351e14","last_reissued_at":"2026-07-05T05:32:31.324944Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:32:31.324944Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Masked Autoencoders that Listen","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","eess.AS"],"primary_cat":"cs.SD","authors_text":"Alexei Baevski, Christoph Feichtenhofer, Florian Metze, Hu Xu, Juncheng Li, Michael Auli, Po-Yao Huang, Wojciech Galuba","submitted_at":"2022-07-13T17:59:55Z","abstract_excerpt":"This paper studies a simple extension of image-based Masked Autoencoders (MAE) to self-supervised representation learning from audio spectrograms. Following the Transformer encoder-decoder design in MAE, our Audio-MAE first encodes audio spectrogram patches with a high masking ratio, feeding only the non-masked tokens through encoder layers. The decoder then re-orders and decodes the encoded context padded with mask tokens, in order to reconstruct the input spectrogram. We find it beneficial to incorporate local window attention in the decoder, as audio spectrograms are highly correlated in lo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2207.06405","kind":"arxiv","version":3},"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/2207.06405/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":"2207.06405","created_at":"2026-07-05T05:32:31.324992+00:00"},{"alias_kind":"arxiv_version","alias_value":"2207.06405v3","created_at":"2026-07-05T05:32:31.324992+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2207.06405","created_at":"2026-07-05T05:32:31.324992+00:00"},{"alias_kind":"pith_short_12","alias_value":"PY4GW77WHJS4","created_at":"2026-07-05T05:32:31.324992+00:00"},{"alias_kind":"pith_short_16","alias_value":"PY4GW77WHJS4TFYI","created_at":"2026-07-05T05:32:31.324992+00:00"},{"alias_kind":"pith_short_8","alias_value":"PY4GW77W","created_at":"2026-07-05T05:32:31.324992+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.09780","citing_title":"Quality-Diversity Search in Sound Generation: Investigating Innovation Engines for Audio Exploration","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2606.21411","citing_title":"CoughPhase-CLR: Designing an acoustics-informed foundation model for coughing sound classification","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2603.03190","citing_title":"Expectation and Acoustic Neural Network Representations Enhance Music Identification from Brain Activity","ref_index":73,"is_internal_anchor":false},{"citing_arxiv_id":"2309.16588","citing_title":"Vision Transformers Need Registers","ref_index":115,"is_internal_anchor":false},{"citing_arxiv_id":"2304.07193","citing_title":"DINOv2: Learning Robust Visual Features without Supervision","ref_index":27,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PY4GW77WHJS4TFYIW4EVYS7WD5","json":"https://pith.science/pith/PY4GW77WHJS4TFYIW4EVYS7WD5.json","graph_json":"https://pith.science/api/pith-number/PY4GW77WHJS4TFYIW4EVYS7WD5/graph.json","events_json":"https://pith.science/api/pith-number/PY4GW77WHJS4TFYIW4EVYS7WD5/events.json","paper":"https://pith.science/paper/PY4GW77W"},"agent_actions":{"view_html":"https://pith.science/pith/PY4GW77WHJS4TFYIW4EVYS7WD5","download_json":"https://pith.science/pith/PY4GW77WHJS4TFYIW4EVYS7WD5.json","view_paper":"https://pith.science/paper/PY4GW77W","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2207.06405&json=true","fetch_graph":"https://pith.science/api/pith-number/PY4GW77WHJS4TFYIW4EVYS7WD5/graph.json","fetch_events":"https://pith.science/api/pith-number/PY4GW77WHJS4TFYIW4EVYS7WD5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PY4GW77WHJS4TFYIW4EVYS7WD5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PY4GW77WHJS4TFYIW4EVYS7WD5/action/storage_attestation","attest_author":"https://pith.science/pith/PY4GW77WHJS4TFYIW4EVYS7WD5/action/author_attestation","sign_citation":"https://pith.science/pith/PY4GW77WHJS4TFYIW4EVYS7WD5/action/citation_signature","submit_replication":"https://pith.science/pith/PY4GW77WHJS4TFYIW4EVYS7WD5/action/replication_record"}},"created_at":"2026-07-05T05:32:31.324992+00:00","updated_at":"2026-07-05T05:32:31.324992+00:00"}