{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:AVMK5JFDXEBMRP3STI4ZU3O5Q3","short_pith_number":"pith:AVMK5JFD","schema_version":"1.0","canonical_sha256":"0558aea4a3b902c8bf729a399a6ddd86f69d3d34f41e8ee399495773b67b4ffc","source":{"kind":"arxiv","id":"2410.04159","version":1},"attestation_state":"computed","paper":{"title":"Efficient and Robust Long-Form Speech Recognition with Hybrid H3-Conformer","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.SD","authors_text":"Shinsuke Sakai, Tatsuya Kawahara, Tomoki Honda","submitted_at":"2024-10-05T13:46:01Z","abstract_excerpt":"Recently, Conformer has achieved state-of-the-art performance in many speech recognition tasks. However, the Transformer-based models show significant deterioration for long-form speech, such as lectures, because the self-attention mechanism becomes unreliable with the computation of the square order of the input length. To solve the problem, we incorporate a kind of state-space model, Hungry Hungry Hippos (H3), to replace or complement the multi-head self-attention (MHSA). H3 allows for efficient modeling of long-form sequences with a linear-order computation. In experiments using two dataset"},"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":"2410.04159","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2024-10-05T13:46:01Z","cross_cats_sorted":["eess.AS"],"title_canon_sha256":"0a8b0d52bea68511d9e38616b632a8484f8f16c21f674083b75fa51a7197ea39","abstract_canon_sha256":"ab09db52d4eac7b55fdbe0688a911b054ebafb1df2a9509f05a2b140061db635"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:16:29.811596Z","signature_b64":"kEDhA7EF3WKpM5L0Ec9E1ELjOHYx3nV620cmXAIQDrxIPvs/PUn2vWfY39YKU4BDY49faS0e7DywdACpjQLgCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0558aea4a3b902c8bf729a399a6ddd86f69d3d34f41e8ee399495773b67b4ffc","last_reissued_at":"2026-07-05T09:16:29.811012Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:16:29.811012Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Efficient and Robust Long-Form Speech Recognition with Hybrid H3-Conformer","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.SD","authors_text":"Shinsuke Sakai, Tatsuya Kawahara, Tomoki Honda","submitted_at":"2024-10-05T13:46:01Z","abstract_excerpt":"Recently, Conformer has achieved state-of-the-art performance in many speech recognition tasks. However, the Transformer-based models show significant deterioration for long-form speech, such as lectures, because the self-attention mechanism becomes unreliable with the computation of the square order of the input length. To solve the problem, we incorporate a kind of state-space model, Hungry Hungry Hippos (H3), to replace or complement the multi-head self-attention (MHSA). H3 allows for efficient modeling of long-form sequences with a linear-order computation. In experiments using two dataset"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.04159","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/2410.04159/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":"2410.04159","created_at":"2026-07-05T09:16:29.811083+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.04159v1","created_at":"2026-07-05T09:16:29.811083+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.04159","created_at":"2026-07-05T09:16:29.811083+00:00"},{"alias_kind":"pith_short_12","alias_value":"AVMK5JFDXEBM","created_at":"2026-07-05T09:16:29.811083+00:00"},{"alias_kind":"pith_short_16","alias_value":"AVMK5JFDXEBMRP3S","created_at":"2026-07-05T09:16:29.811083+00:00"},{"alias_kind":"pith_short_8","alias_value":"AVMK5JFD","created_at":"2026-07-05T09:16:29.811083+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.07355","citing_title":"TTF: Temporal Token Fusion for Efficient Video-Language Model","ref_index":20,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AVMK5JFDXEBMRP3STI4ZU3O5Q3","json":"https://pith.science/pith/AVMK5JFDXEBMRP3STI4ZU3O5Q3.json","graph_json":"https://pith.science/api/pith-number/AVMK5JFDXEBMRP3STI4ZU3O5Q3/graph.json","events_json":"https://pith.science/api/pith-number/AVMK5JFDXEBMRP3STI4ZU3O5Q3/events.json","paper":"https://pith.science/paper/AVMK5JFD"},"agent_actions":{"view_html":"https://pith.science/pith/AVMK5JFDXEBMRP3STI4ZU3O5Q3","download_json":"https://pith.science/pith/AVMK5JFDXEBMRP3STI4ZU3O5Q3.json","view_paper":"https://pith.science/paper/AVMK5JFD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.04159&json=true","fetch_graph":"https://pith.science/api/pith-number/AVMK5JFDXEBMRP3STI4ZU3O5Q3/graph.json","fetch_events":"https://pith.science/api/pith-number/AVMK5JFDXEBMRP3STI4ZU3O5Q3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AVMK5JFDXEBMRP3STI4ZU3O5Q3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AVMK5JFDXEBMRP3STI4ZU3O5Q3/action/storage_attestation","attest_author":"https://pith.science/pith/AVMK5JFDXEBMRP3STI4ZU3O5Q3/action/author_attestation","sign_citation":"https://pith.science/pith/AVMK5JFDXEBMRP3STI4ZU3O5Q3/action/citation_signature","submit_replication":"https://pith.science/pith/AVMK5JFDXEBMRP3STI4ZU3O5Q3/action/replication_record"}},"created_at":"2026-07-05T09:16:29.811083+00:00","updated_at":"2026-07-05T09:16:29.811083+00:00"}