{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:NSJECOXPAPCYSZUHCFJLEMQGQW","short_pith_number":"pith:NSJECOXP","schema_version":"1.0","canonical_sha256":"6c92413aef03c58966871152b232068594e5a196cc0efcbe75bab75abdc1c905","source":{"kind":"arxiv","id":"2210.17079","version":1},"attestation_state":"computed","paper":{"title":"FusionFormer: Fusing Operations in Transformer for Efficient Streaming Speech Recognition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","eess.AS"],"primary_cat":"cs.SD","authors_text":"Binbin Zhang, Changbao Zhu, Di Wu, Dongfang Li, Fuping Pan, Pengshen Zhang, Wenpeng Li, Xingchen Song, Zhendong Peng, Zhiyong Wu, Zhongqin Wu","submitted_at":"2022-10-31T06:01:02Z","abstract_excerpt":"The recently proposed Conformer architecture which combines convolution with attention to capture both local and global dependencies has become the \\textit{de facto} backbone model for Automatic Speech Recognition~(ASR). Inherited from the Natural Language Processing (NLP) tasks, the architecture takes Layer Normalization~(LN) as a default normalization technique. However, through a series of systematic studies, we find that LN might take 10\\% of the inference time despite that it only contributes to 0.1\\% of the FLOPs. This motivates us to replace LN with other normalization techniques, e.g.,"},"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":"2210.17079","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2022-10-31T06:01:02Z","cross_cats_sorted":["cs.CL","eess.AS"],"title_canon_sha256":"602a2aa256ca4c18de63bf286e088c7ff2cde61e69e8defc42c51096e6483ccd","abstract_canon_sha256":"9d4a8e04d4d4277af2a84f342a4eaa062b7060969549810414579f235668af23"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:11:54.762446Z","signature_b64":"AxYP+HXo8Le9CWjNprW6dvpDeB1yUs9tap/Nz/ug8YYb7cbz3iWIrAmr65eWj2KVmThe67UyKx5Z1xKUzx4nBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6c92413aef03c58966871152b232068594e5a196cc0efcbe75bab75abdc1c905","last_reissued_at":"2026-07-05T05:11:54.762001Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:11:54.762001Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"FusionFormer: Fusing Operations in Transformer for Efficient Streaming Speech Recognition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","eess.AS"],"primary_cat":"cs.SD","authors_text":"Binbin Zhang, Changbao Zhu, Di Wu, Dongfang Li, Fuping Pan, Pengshen Zhang, Wenpeng Li, Xingchen Song, Zhendong Peng, Zhiyong Wu, Zhongqin Wu","submitted_at":"2022-10-31T06:01:02Z","abstract_excerpt":"The recently proposed Conformer architecture which combines convolution with attention to capture both local and global dependencies has become the \\textit{de facto} backbone model for Automatic Speech Recognition~(ASR). Inherited from the Natural Language Processing (NLP) tasks, the architecture takes Layer Normalization~(LN) as a default normalization technique. However, through a series of systematic studies, we find that LN might take 10\\% of the inference time despite that it only contributes to 0.1\\% of the FLOPs. This motivates us to replace LN with other normalization techniques, e.g.,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.17079","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/2210.17079/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":"2210.17079","created_at":"2026-07-05T05:11:54.762073+00:00"},{"alias_kind":"arxiv_version","alias_value":"2210.17079v1","created_at":"2026-07-05T05:11:54.762073+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.17079","created_at":"2026-07-05T05:11:54.762073+00:00"},{"alias_kind":"pith_short_12","alias_value":"NSJECOXPAPCY","created_at":"2026-07-05T05:11:54.762073+00:00"},{"alias_kind":"pith_short_16","alias_value":"NSJECOXPAPCYSZUH","created_at":"2026-07-05T05:11:54.762073+00:00"},{"alias_kind":"pith_short_8","alias_value":"NSJECOXP","created_at":"2026-07-05T05:11:54.762073+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NSJECOXPAPCYSZUHCFJLEMQGQW","json":"https://pith.science/pith/NSJECOXPAPCYSZUHCFJLEMQGQW.json","graph_json":"https://pith.science/api/pith-number/NSJECOXPAPCYSZUHCFJLEMQGQW/graph.json","events_json":"https://pith.science/api/pith-number/NSJECOXPAPCYSZUHCFJLEMQGQW/events.json","paper":"https://pith.science/paper/NSJECOXP"},"agent_actions":{"view_html":"https://pith.science/pith/NSJECOXPAPCYSZUHCFJLEMQGQW","download_json":"https://pith.science/pith/NSJECOXPAPCYSZUHCFJLEMQGQW.json","view_paper":"https://pith.science/paper/NSJECOXP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2210.17079&json=true","fetch_graph":"https://pith.science/api/pith-number/NSJECOXPAPCYSZUHCFJLEMQGQW/graph.json","fetch_events":"https://pith.science/api/pith-number/NSJECOXPAPCYSZUHCFJLEMQGQW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NSJECOXPAPCYSZUHCFJLEMQGQW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NSJECOXPAPCYSZUHCFJLEMQGQW/action/storage_attestation","attest_author":"https://pith.science/pith/NSJECOXPAPCYSZUHCFJLEMQGQW/action/author_attestation","sign_citation":"https://pith.science/pith/NSJECOXPAPCYSZUHCFJLEMQGQW/action/citation_signature","submit_replication":"https://pith.science/pith/NSJECOXPAPCYSZUHCFJLEMQGQW/action/replication_record"}},"created_at":"2026-07-05T05:11:54.762073+00:00","updated_at":"2026-07-05T05:11:54.762073+00:00"}