{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:EOQPCTCFL2JNJ3U4OB3L3PPB2M","short_pith_number":"pith:EOQPCTCF","schema_version":"1.0","canonical_sha256":"23a0f14c455e92d4ee9c7076bdbde1d3069242da39428cdc524264954ab4205a","source":{"kind":"arxiv","id":"2103.06089","version":1},"attestation_state":"computed","paper":{"title":"Variable-rate discrete representation learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.SD","eess.AS"],"primary_cat":"cs.LG","authors_text":"Charlie Nash, Jesse Engel, Karen Simonyan, Sander Dieleman","submitted_at":"2021-03-10T14:42:31Z","abstract_excerpt":"Semantically meaningful information content in perceptual signals is usually unevenly distributed. In speech signals for example, there are often many silences, and the speed of pronunciation can vary considerably. In this work, we propose slow autoencoders (SlowAEs) for unsupervised learning of high-level variable-rate discrete representations of sequences, and apply them to speech. We show that the resulting event-based representations automatically grow or shrink depending on the density of salient information in the input signals, while still allowing for faithful signal reconstruction. We"},"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":"2103.06089","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-03-10T14:42:31Z","cross_cats_sorted":["cs.CL","cs.SD","eess.AS"],"title_canon_sha256":"f53d5dc7ef6bb01194f37fdd29b68afd47f8bec076db9c7c510de6feb8c26e31","abstract_canon_sha256":"d626bf439790f242e833a9e8f04416305b0a6b18401655591bf1a7260d161b31"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:22:01.479243Z","signature_b64":"FTuuGFNEbuC6RQRSlcmRqJ606GdMp0Oo7ggIJJcg77/78YRXb2ZP+Qd4YFv2GjNmQkGXC5TlcOpi8va0OE2sCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"23a0f14c455e92d4ee9c7076bdbde1d3069242da39428cdc524264954ab4205a","last_reissued_at":"2026-07-05T02:22:01.478727Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:22:01.478727Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Variable-rate discrete representation learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.SD","eess.AS"],"primary_cat":"cs.LG","authors_text":"Charlie Nash, Jesse Engel, Karen Simonyan, Sander Dieleman","submitted_at":"2021-03-10T14:42:31Z","abstract_excerpt":"Semantically meaningful information content in perceptual signals is usually unevenly distributed. In speech signals for example, there are often many silences, and the speed of pronunciation can vary considerably. In this work, we propose slow autoencoders (SlowAEs) for unsupervised learning of high-level variable-rate discrete representations of sequences, and apply them to speech. We show that the resulting event-based representations automatically grow or shrink depending on the density of salient information in the input signals, while still allowing for faithful signal reconstruction. We"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2103.06089","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/2103.06089/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":"2103.06089","created_at":"2026-07-05T02:22:01.478790+00:00"},{"alias_kind":"arxiv_version","alias_value":"2103.06089v1","created_at":"2026-07-05T02:22:01.478790+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2103.06089","created_at":"2026-07-05T02:22:01.478790+00:00"},{"alias_kind":"pith_short_12","alias_value":"EOQPCTCFL2JN","created_at":"2026-07-05T02:22:01.478790+00:00"},{"alias_kind":"pith_short_16","alias_value":"EOQPCTCFL2JNJ3U4","created_at":"2026-07-05T02:22:01.478790+00:00"},{"alias_kind":"pith_short_8","alias_value":"EOQPCTCF","created_at":"2026-07-05T02:22:01.478790+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.31247","citing_title":"FlexiSLM: A Dynamic and Controllable Frame Rate Spoken Language Model","ref_index":139,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29480","citing_title":"DTM-Codec: Dynamic Token Masking for VFR Speech Coding with Efficient Boundary Selection","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2211.15089","citing_title":"Continuous diffusion for categorical data","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2309.15505","citing_title":"Finite Scalar Quantization: VQ-VAE Made Simple","ref_index":9,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EOQPCTCFL2JNJ3U4OB3L3PPB2M","json":"https://pith.science/pith/EOQPCTCFL2JNJ3U4OB3L3PPB2M.json","graph_json":"https://pith.science/api/pith-number/EOQPCTCFL2JNJ3U4OB3L3PPB2M/graph.json","events_json":"https://pith.science/api/pith-number/EOQPCTCFL2JNJ3U4OB3L3PPB2M/events.json","paper":"https://pith.science/paper/EOQPCTCF"},"agent_actions":{"view_html":"https://pith.science/pith/EOQPCTCFL2JNJ3U4OB3L3PPB2M","download_json":"https://pith.science/pith/EOQPCTCFL2JNJ3U4OB3L3PPB2M.json","view_paper":"https://pith.science/paper/EOQPCTCF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2103.06089&json=true","fetch_graph":"https://pith.science/api/pith-number/EOQPCTCFL2JNJ3U4OB3L3PPB2M/graph.json","fetch_events":"https://pith.science/api/pith-number/EOQPCTCFL2JNJ3U4OB3L3PPB2M/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EOQPCTCFL2JNJ3U4OB3L3PPB2M/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EOQPCTCFL2JNJ3U4OB3L3PPB2M/action/storage_attestation","attest_author":"https://pith.science/pith/EOQPCTCFL2JNJ3U4OB3L3PPB2M/action/author_attestation","sign_citation":"https://pith.science/pith/EOQPCTCFL2JNJ3U4OB3L3PPB2M/action/citation_signature","submit_replication":"https://pith.science/pith/EOQPCTCFL2JNJ3U4OB3L3PPB2M/action/replication_record"}},"created_at":"2026-07-05T02:22:01.478790+00:00","updated_at":"2026-07-05T02:22:01.478790+00:00"}