{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:4EWXGI7VZDAZ2ZYMH3AKVGMBHD","short_pith_number":"pith:4EWXGI7V","schema_version":"1.0","canonical_sha256":"e12d7323f5c8c19d670c3ec0aa998138f93fbc99057c3f98c16da10f3f2c9ff2","source":{"kind":"arxiv","id":"2103.15722","version":4},"attestation_state":"computed","paper":{"title":"Transformer-based end-to-end speech recognition with residual Gaussian-based self-attention","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG","eess.AS"],"primary_cat":"cs.SD","authors_text":"Chengdong Liang, Menglong Xu, Xiao-Lei Zhang","submitted_at":"2021-03-29T16:09:00Z","abstract_excerpt":"Self-attention (SA), which encodes vector sequences according to their pairwise similarity, is widely used in speech recognition due to its strong context modeling ability. However, when applied to long sequence data, its accuracy is reduced. This is caused by the fact that its weighted average operator may lead to the dispersion of the attention distribution, which results in the relationship between adjacent signals ignored. To address this issue, in this paper, we introduce relative-position-awareness self-attention (RPSA). It not only maintains the global-range dependency modeling ability "},"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":"2103.15722","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2021-03-29T16:09:00Z","cross_cats_sorted":["cs.CL","cs.LG","eess.AS"],"title_canon_sha256":"e2655abee3165115df592496d37ed470e1fdb6e3a49028f030781082ed75203e","abstract_canon_sha256":"32bebed15b321e25f2f9be8e63c778335e3dc64c40f499a67b6100f15631a82b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:20:54.855362Z","signature_b64":"LnbyIgljQl2Z/ZKNqi57MI6KLzFZ7UBbWCvJNLu9DlPEPhCd9ETM6n/CTT/saYeo+utapjeV4U0oj+219pVXAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e12d7323f5c8c19d670c3ec0aa998138f93fbc99057c3f98c16da10f3f2c9ff2","last_reissued_at":"2026-07-05T03:20:54.854870Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:20:54.854870Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Transformer-based end-to-end speech recognition with residual Gaussian-based self-attention","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG","eess.AS"],"primary_cat":"cs.SD","authors_text":"Chengdong Liang, Menglong Xu, Xiao-Lei Zhang","submitted_at":"2021-03-29T16:09:00Z","abstract_excerpt":"Self-attention (SA), which encodes vector sequences according to their pairwise similarity, is widely used in speech recognition due to its strong context modeling ability. However, when applied to long sequence data, its accuracy is reduced. This is caused by the fact that its weighted average operator may lead to the dispersion of the attention distribution, which results in the relationship between adjacent signals ignored. To address this issue, in this paper, we introduce relative-position-awareness self-attention (RPSA). It not only maintains the global-range dependency modeling ability "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2103.15722","kind":"arxiv","version":4},"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/2103.15722/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":"2103.15722","created_at":"2026-07-05T03:20:54.854928+00:00"},{"alias_kind":"arxiv_version","alias_value":"2103.15722v4","created_at":"2026-07-05T03:20:54.854928+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2103.15722","created_at":"2026-07-05T03:20:54.854928+00:00"},{"alias_kind":"pith_short_12","alias_value":"4EWXGI7VZDAZ","created_at":"2026-07-05T03:20:54.854928+00:00"},{"alias_kind":"pith_short_16","alias_value":"4EWXGI7VZDAZ2ZYM","created_at":"2026-07-05T03:20:54.854928+00:00"},{"alias_kind":"pith_short_8","alias_value":"4EWXGI7V","created_at":"2026-07-05T03:20:54.854928+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/4EWXGI7VZDAZ2ZYMH3AKVGMBHD","json":"https://pith.science/pith/4EWXGI7VZDAZ2ZYMH3AKVGMBHD.json","graph_json":"https://pith.science/api/pith-number/4EWXGI7VZDAZ2ZYMH3AKVGMBHD/graph.json","events_json":"https://pith.science/api/pith-number/4EWXGI7VZDAZ2ZYMH3AKVGMBHD/events.json","paper":"https://pith.science/paper/4EWXGI7V"},"agent_actions":{"view_html":"https://pith.science/pith/4EWXGI7VZDAZ2ZYMH3AKVGMBHD","download_json":"https://pith.science/pith/4EWXGI7VZDAZ2ZYMH3AKVGMBHD.json","view_paper":"https://pith.science/paper/4EWXGI7V","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2103.15722&json=true","fetch_graph":"https://pith.science/api/pith-number/4EWXGI7VZDAZ2ZYMH3AKVGMBHD/graph.json","fetch_events":"https://pith.science/api/pith-number/4EWXGI7VZDAZ2ZYMH3AKVGMBHD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4EWXGI7VZDAZ2ZYMH3AKVGMBHD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4EWXGI7VZDAZ2ZYMH3AKVGMBHD/action/storage_attestation","attest_author":"https://pith.science/pith/4EWXGI7VZDAZ2ZYMH3AKVGMBHD/action/author_attestation","sign_citation":"https://pith.science/pith/4EWXGI7VZDAZ2ZYMH3AKVGMBHD/action/citation_signature","submit_replication":"https://pith.science/pith/4EWXGI7VZDAZ2ZYMH3AKVGMBHD/action/replication_record"}},"created_at":"2026-07-05T03:20:54.854928+00:00","updated_at":"2026-07-05T03:20:54.854928+00:00"}