{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:BIM3BNVMK5BHSYPKOUQM672MAI","short_pith_number":"pith:BIM3BNVM","schema_version":"1.0","canonical_sha256":"0a19b0b6ac57427961ea7520cf7f4c0220bcc3ce48cb61992d34a7956acc494c","source":{"kind":"arxiv","id":"2012.15150","version":2},"attestation_state":"computed","paper":{"title":"Improving BERT with Syntax-aware Local Attention","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chao Li, Ke Xu, Qingyu Zhou, Yunbo Cao, Zhongli Li","submitted_at":"2020-12-30T13:29:58Z","abstract_excerpt":"Pre-trained Transformer-based neural language models, such as BERT, have achieved remarkable results on varieties of NLP tasks. Recent works have shown that attention-based models can benefit from more focused attention over local regions. Most of them restrict the attention scope within a linear span, or confine to certain tasks such as machine translation and question answering. In this paper, we propose a syntax-aware local attention, where the attention scopes are restrained based on the distances in the syntactic structure. The proposed syntax-aware local attention can be integrated with "},"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":"2012.15150","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2020-12-30T13:29:58Z","cross_cats_sorted":[],"title_canon_sha256":"54298f9eb6cb724872747eaf879da8acad3a0e1527654bd45cccce02bb90a448","abstract_canon_sha256":"48c05af228cfd21b858096e9f4406e32559a1a3c468ccda16277b58166c02631"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:42:28.737867Z","signature_b64":"6DgIivvDMNTd0X91QU2Uycg6qylsi9bs5rSuFTzBGDO/9Npzh93xXTOZqhfRAw7rXC4O9ueHj9mhbOV1XtsaCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0a19b0b6ac57427961ea7520cf7f4c0220bcc3ce48cb61992d34a7956acc494c","last_reissued_at":"2026-07-05T02:42:28.737386Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:42:28.737386Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving BERT with Syntax-aware Local Attention","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chao Li, Ke Xu, Qingyu Zhou, Yunbo Cao, Zhongli Li","submitted_at":"2020-12-30T13:29:58Z","abstract_excerpt":"Pre-trained Transformer-based neural language models, such as BERT, have achieved remarkable results on varieties of NLP tasks. Recent works have shown that attention-based models can benefit from more focused attention over local regions. Most of them restrict the attention scope within a linear span, or confine to certain tasks such as machine translation and question answering. In this paper, we propose a syntax-aware local attention, where the attention scopes are restrained based on the distances in the syntactic structure. The proposed syntax-aware local attention can be integrated with "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2012.15150","kind":"arxiv","version":2},"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/2012.15150/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":"2012.15150","created_at":"2026-07-05T02:42:28.737447+00:00"},{"alias_kind":"arxiv_version","alias_value":"2012.15150v2","created_at":"2026-07-05T02:42:28.737447+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2012.15150","created_at":"2026-07-05T02:42:28.737447+00:00"},{"alias_kind":"pith_short_12","alias_value":"BIM3BNVMK5BH","created_at":"2026-07-05T02:42:28.737447+00:00"},{"alias_kind":"pith_short_16","alias_value":"BIM3BNVMK5BHSYPK","created_at":"2026-07-05T02:42:28.737447+00:00"},{"alias_kind":"pith_short_8","alias_value":"BIM3BNVM","created_at":"2026-07-05T02:42:28.737447+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/BIM3BNVMK5BHSYPKOUQM672MAI","json":"https://pith.science/pith/BIM3BNVMK5BHSYPKOUQM672MAI.json","graph_json":"https://pith.science/api/pith-number/BIM3BNVMK5BHSYPKOUQM672MAI/graph.json","events_json":"https://pith.science/api/pith-number/BIM3BNVMK5BHSYPKOUQM672MAI/events.json","paper":"https://pith.science/paper/BIM3BNVM"},"agent_actions":{"view_html":"https://pith.science/pith/BIM3BNVMK5BHSYPKOUQM672MAI","download_json":"https://pith.science/pith/BIM3BNVMK5BHSYPKOUQM672MAI.json","view_paper":"https://pith.science/paper/BIM3BNVM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2012.15150&json=true","fetch_graph":"https://pith.science/api/pith-number/BIM3BNVMK5BHSYPKOUQM672MAI/graph.json","fetch_events":"https://pith.science/api/pith-number/BIM3BNVMK5BHSYPKOUQM672MAI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BIM3BNVMK5BHSYPKOUQM672MAI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BIM3BNVMK5BHSYPKOUQM672MAI/action/storage_attestation","attest_author":"https://pith.science/pith/BIM3BNVMK5BHSYPKOUQM672MAI/action/author_attestation","sign_citation":"https://pith.science/pith/BIM3BNVMK5BHSYPKOUQM672MAI/action/citation_signature","submit_replication":"https://pith.science/pith/BIM3BNVMK5BHSYPKOUQM672MAI/action/replication_record"}},"created_at":"2026-07-05T02:42:28.737447+00:00","updated_at":"2026-07-05T02:42:28.737447+00:00"}