{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:A2ZU5K2TGDRMSFB5LWMIX6TXUJ","short_pith_number":"pith:A2ZU5K2T","schema_version":"1.0","canonical_sha256":"06b34eab5330e2c9143d5d988bfa77a251d80f87e8a8edc781716df3380ef8de","source":{"kind":"arxiv","id":"2507.02915","version":1},"attestation_state":"computed","paper":{"title":"Audio-JEPA: Joint-Embedding Predictive Architecture for Audio Representation Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","eess.AS","eess.SP"],"primary_cat":"cs.SD","authors_text":"Emmanouil Benetos (QMUL), Etienne Labb\\'e (IRIT-SAMoVA), Ludovic Tuncay (IRIT-SAMoVA), Thomas Pellegrini (IRIT-SAMoVA)","submitted_at":"2025-06-25T08:38:27Z","abstract_excerpt":"Building on the Joint-Embedding Predictive Architecture (JEPA) paradigm, a recent self-supervised learning framework that predicts latent representations of masked regions in high-level feature spaces, we propose Audio-JEPA (Audio Joint-Embedding Predictive Architecture), tailored specifically for audio data. Audio-JEPA uses a simple Vision Transformer backbone to predict latent representations of masked spectrogram patches rather than reconstructing raw audio. We pre-train on unlabeled AudioSet clips (10s, 32kHz) with random patch masking on mel-spectrograms. We evaluate on the X-ARES suite c"},"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":"2507.02915","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2025-06-25T08:38:27Z","cross_cats_sorted":["cs.AI","cs.LG","eess.AS","eess.SP"],"title_canon_sha256":"e4594f6a56d95835967af77fb0979ff02598b9d8abf9514fea4523c8f4299e9d","abstract_canon_sha256":"a43f2a91c31acface75d3f371bee55c15cdb67a422055dc40abd5a2884ac8975"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:31:48.642541Z","signature_b64":"AoDSzSAf77IvQ5/gKzeLPD36aRqwxCfyW8bkl9HrupJAcBgI1oYGLq8aUwhU3Q/6Envm0qqjBf7MYlKz1oOiCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"06b34eab5330e2c9143d5d988bfa77a251d80f87e8a8edc781716df3380ef8de","last_reissued_at":"2026-07-05T11:31:48.641916Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:31:48.641916Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Audio-JEPA: Joint-Embedding Predictive Architecture for Audio Representation Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","eess.AS","eess.SP"],"primary_cat":"cs.SD","authors_text":"Emmanouil Benetos (QMUL), Etienne Labb\\'e (IRIT-SAMoVA), Ludovic Tuncay (IRIT-SAMoVA), Thomas Pellegrini (IRIT-SAMoVA)","submitted_at":"2025-06-25T08:38:27Z","abstract_excerpt":"Building on the Joint-Embedding Predictive Architecture (JEPA) paradigm, a recent self-supervised learning framework that predicts latent representations of masked regions in high-level feature spaces, we propose Audio-JEPA (Audio Joint-Embedding Predictive Architecture), tailored specifically for audio data. Audio-JEPA uses a simple Vision Transformer backbone to predict latent representations of masked spectrogram patches rather than reconstructing raw audio. We pre-train on unlabeled AudioSet clips (10s, 32kHz) with random patch masking on mel-spectrograms. We evaluate on the X-ARES suite c"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.02915","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/2507.02915/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":"2507.02915","created_at":"2026-07-05T11:31:48.641984+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.02915v1","created_at":"2026-07-05T11:31:48.641984+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.02915","created_at":"2026-07-05T11:31:48.641984+00:00"},{"alias_kind":"pith_short_12","alias_value":"A2ZU5K2TGDRM","created_at":"2026-07-05T11:31:48.641984+00:00"},{"alias_kind":"pith_short_16","alias_value":"A2ZU5K2TGDRMSFB5","created_at":"2026-07-05T11:31:48.641984+00:00"},{"alias_kind":"pith_short_8","alias_value":"A2ZU5K2T","created_at":"2026-07-05T11:31:48.641984+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.25713","citing_title":"Frequency-Aware Self-Supervised Music Representation Learning","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01909","citing_title":"Echo: A Joint-Embedding Predictive Architecture for Speaker Diarization and Speech Recognition in a Shared Latent Space","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2606.25713","citing_title":"Frequency-Aware Self-Supervised Music Representation Learning","ref_index":32,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/A2ZU5K2TGDRMSFB5LWMIX6TXUJ","json":"https://pith.science/pith/A2ZU5K2TGDRMSFB5LWMIX6TXUJ.json","graph_json":"https://pith.science/api/pith-number/A2ZU5K2TGDRMSFB5LWMIX6TXUJ/graph.json","events_json":"https://pith.science/api/pith-number/A2ZU5K2TGDRMSFB5LWMIX6TXUJ/events.json","paper":"https://pith.science/paper/A2ZU5K2T"},"agent_actions":{"view_html":"https://pith.science/pith/A2ZU5K2TGDRMSFB5LWMIX6TXUJ","download_json":"https://pith.science/pith/A2ZU5K2TGDRMSFB5LWMIX6TXUJ.json","view_paper":"https://pith.science/paper/A2ZU5K2T","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.02915&json=true","fetch_graph":"https://pith.science/api/pith-number/A2ZU5K2TGDRMSFB5LWMIX6TXUJ/graph.json","fetch_events":"https://pith.science/api/pith-number/A2ZU5K2TGDRMSFB5LWMIX6TXUJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/A2ZU5K2TGDRMSFB5LWMIX6TXUJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/A2ZU5K2TGDRMSFB5LWMIX6TXUJ/action/storage_attestation","attest_author":"https://pith.science/pith/A2ZU5K2TGDRMSFB5LWMIX6TXUJ/action/author_attestation","sign_citation":"https://pith.science/pith/A2ZU5K2TGDRMSFB5LWMIX6TXUJ/action/citation_signature","submit_replication":"https://pith.science/pith/A2ZU5K2TGDRMSFB5LWMIX6TXUJ/action/replication_record"}},"created_at":"2026-07-05T11:31:48.641984+00:00","updated_at":"2026-07-05T11:31:48.641984+00:00"}