{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:ORHH5MCRFO7MME3QRFUIDNU2H5","short_pith_number":"pith:ORHH5MCR","schema_version":"1.0","canonical_sha256":"744e7eb0512bbec61370896881b69a3f6cb06a17cd45f21ae7cc81ec50a8b73e","source":{"kind":"arxiv","id":"2305.18281","version":1},"attestation_state":"computed","paper":{"title":"HyperConformer: Multi-head HyperMixer for Efficient Speech Recognition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","eess.AS"],"primary_cat":"cs.CL","authors_text":"Florian Mai, Juan Zuluaga-Gomez, Petr Motlicek, Titouan Parcollet","submitted_at":"2023-05-29T17:53:04Z","abstract_excerpt":"State-of-the-art ASR systems have achieved promising results by modeling local and global interactions separately. While the former can be computed efficiently, global interactions are usually modeled via attention mechanisms, which are expensive for long input sequences. Here, we address this by extending HyperMixer, an efficient alternative to attention exhibiting linear complexity, to the Conformer architecture for speech recognition, leading to HyperConformer. In particular, multi-head HyperConformer achieves comparable or higher recognition performance while being more efficient than Conf"},"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":"2305.18281","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-29T17:53:04Z","cross_cats_sorted":["cs.AI","cs.LG","eess.AS"],"title_canon_sha256":"281680186f58301bbe3b0d21761512e21ff462e36506c3f0e3770eac604efeec","abstract_canon_sha256":"cbfae4211748cc1f122e457c7c2fb64c67f2df6a8c64baf22cd6e1d7534a31ee"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:14:54.501034Z","signature_b64":"Z3YAvaLcVpPa+1nm+j0zAf6n/sBC2c34AzehuqnuI7KxgFvuoLC2vkvEHz2HB0TM3lrQ98uo4weS9U1GJ8JeDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"744e7eb0512bbec61370896881b69a3f6cb06a17cd45f21ae7cc81ec50a8b73e","last_reissued_at":"2026-07-05T06:14:54.500599Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:14:54.500599Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"HyperConformer: Multi-head HyperMixer for Efficient Speech Recognition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","eess.AS"],"primary_cat":"cs.CL","authors_text":"Florian Mai, Juan Zuluaga-Gomez, Petr Motlicek, Titouan Parcollet","submitted_at":"2023-05-29T17:53:04Z","abstract_excerpt":"State-of-the-art ASR systems have achieved promising results by modeling local and global interactions separately. While the former can be computed efficiently, global interactions are usually modeled via attention mechanisms, which are expensive for long input sequences. Here, we address this by extending HyperMixer, an efficient alternative to attention exhibiting linear complexity, to the Conformer architecture for speech recognition, leading to HyperConformer. In particular, multi-head HyperConformer achieves comparable or higher recognition performance while being more efficient than Conf"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.18281","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/2305.18281/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":"2305.18281","created_at":"2026-07-05T06:14:54.500654+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.18281v1","created_at":"2026-07-05T06:14:54.500654+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.18281","created_at":"2026-07-05T06:14:54.500654+00:00"},{"alias_kind":"pith_short_12","alias_value":"ORHH5MCRFO7M","created_at":"2026-07-05T06:14:54.500654+00:00"},{"alias_kind":"pith_short_16","alias_value":"ORHH5MCRFO7MME3Q","created_at":"2026-07-05T06:14:54.500654+00:00"},{"alias_kind":"pith_short_8","alias_value":"ORHH5MCR","created_at":"2026-07-05T06:14:54.500654+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":73,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ORHH5MCRFO7MME3QRFUIDNU2H5","json":"https://pith.science/pith/ORHH5MCRFO7MME3QRFUIDNU2H5.json","graph_json":"https://pith.science/api/pith-number/ORHH5MCRFO7MME3QRFUIDNU2H5/graph.json","events_json":"https://pith.science/api/pith-number/ORHH5MCRFO7MME3QRFUIDNU2H5/events.json","paper":"https://pith.science/paper/ORHH5MCR"},"agent_actions":{"view_html":"https://pith.science/pith/ORHH5MCRFO7MME3QRFUIDNU2H5","download_json":"https://pith.science/pith/ORHH5MCRFO7MME3QRFUIDNU2H5.json","view_paper":"https://pith.science/paper/ORHH5MCR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.18281&json=true","fetch_graph":"https://pith.science/api/pith-number/ORHH5MCRFO7MME3QRFUIDNU2H5/graph.json","fetch_events":"https://pith.science/api/pith-number/ORHH5MCRFO7MME3QRFUIDNU2H5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ORHH5MCRFO7MME3QRFUIDNU2H5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ORHH5MCRFO7MME3QRFUIDNU2H5/action/storage_attestation","attest_author":"https://pith.science/pith/ORHH5MCRFO7MME3QRFUIDNU2H5/action/author_attestation","sign_citation":"https://pith.science/pith/ORHH5MCRFO7MME3QRFUIDNU2H5/action/citation_signature","submit_replication":"https://pith.science/pith/ORHH5MCRFO7MME3QRFUIDNU2H5/action/replication_record"}},"created_at":"2026-07-05T06:14:54.500654+00:00","updated_at":"2026-07-05T06:14:54.500654+00:00"}