{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:PKNHJTZVKSWBUKDVPLSYS3Q2LX","short_pith_number":"pith:PKNHJTZV","schema_version":"1.0","canonical_sha256":"7a9a74cf3554ac1a28757ae5896e1a5dc2bfce94c3031c7dd242035d6bc733cf","source":{"kind":"arxiv","id":"1911.03875","version":3},"attestation_state":"computed","paper":{"title":"Rethinking Self-Attention: Towards Interpretability in Neural Parsing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Franck Dernoncourt, Khalil Mrini, Ndapa Nakashole, Quan Tran, Trung Bui, Walter Chang","submitted_at":"2019-11-10T08:17:11Z","abstract_excerpt":"Attention mechanisms have improved the performance of NLP tasks while allowing models to remain explainable. Self-attention is currently widely used, however interpretability is difficult due to the numerous attention distributions. Recent work has shown that model representations can benefit from label-specific information, while facilitating interpretation of predictions. We introduce the Label Attention Layer: a new form of self-attention where attention heads represent labels. We test our novel layer by running constituency and dependency parsing experiments and show our new model obtains "},"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":"1911.03875","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-11-10T08:17:11Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"89aa2b9cb4230733592bb97665d790f5d08928d4a135fce39763516f030b4671","abstract_canon_sha256":"926504929588fda1de0aff603821907e3ed5c6c4109d4d77a744baf67e911541"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:47:29.250371Z","signature_b64":"DIFvQtXxgw43H2qamo6CvvnI9c0JT3+DXfCoaEFYgtYACD16eInLfQCHCyfTsgQZNK1ZQzsXM5uT+da4H9PrDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7a9a74cf3554ac1a28757ae5896e1a5dc2bfce94c3031c7dd242035d6bc733cf","last_reissued_at":"2026-07-05T01:47:29.249959Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:47:29.249959Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Rethinking Self-Attention: Towards Interpretability in Neural Parsing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Franck Dernoncourt, Khalil Mrini, Ndapa Nakashole, Quan Tran, Trung Bui, Walter Chang","submitted_at":"2019-11-10T08:17:11Z","abstract_excerpt":"Attention mechanisms have improved the performance of NLP tasks while allowing models to remain explainable. Self-attention is currently widely used, however interpretability is difficult due to the numerous attention distributions. Recent work has shown that model representations can benefit from label-specific information, while facilitating interpretation of predictions. We introduce the Label Attention Layer: a new form of self-attention where attention heads represent labels. We test our novel layer by running constituency and dependency parsing experiments and show our new model obtains "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1911.03875","kind":"arxiv","version":3},"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/1911.03875/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":"1911.03875","created_at":"2026-07-05T01:47:29.250014+00:00"},{"alias_kind":"arxiv_version","alias_value":"1911.03875v3","created_at":"2026-07-05T01:47:29.250014+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1911.03875","created_at":"2026-07-05T01:47:29.250014+00:00"},{"alias_kind":"pith_short_12","alias_value":"PKNHJTZVKSWB","created_at":"2026-07-05T01:47:29.250014+00:00"},{"alias_kind":"pith_short_16","alias_value":"PKNHJTZVKSWBUKDV","created_at":"2026-07-05T01:47:29.250014+00:00"},{"alias_kind":"pith_short_8","alias_value":"PKNHJTZV","created_at":"2026-07-05T01:47:29.250014+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.20453","citing_title":"Your Attention Matters: to Improve Model Robustness to Noise and Spurious Correlations","ref_index":25,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PKNHJTZVKSWBUKDVPLSYS3Q2LX","json":"https://pith.science/pith/PKNHJTZVKSWBUKDVPLSYS3Q2LX.json","graph_json":"https://pith.science/api/pith-number/PKNHJTZVKSWBUKDVPLSYS3Q2LX/graph.json","events_json":"https://pith.science/api/pith-number/PKNHJTZVKSWBUKDVPLSYS3Q2LX/events.json","paper":"https://pith.science/paper/PKNHJTZV"},"agent_actions":{"view_html":"https://pith.science/pith/PKNHJTZVKSWBUKDVPLSYS3Q2LX","download_json":"https://pith.science/pith/PKNHJTZVKSWBUKDVPLSYS3Q2LX.json","view_paper":"https://pith.science/paper/PKNHJTZV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1911.03875&json=true","fetch_graph":"https://pith.science/api/pith-number/PKNHJTZVKSWBUKDVPLSYS3Q2LX/graph.json","fetch_events":"https://pith.science/api/pith-number/PKNHJTZVKSWBUKDVPLSYS3Q2LX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PKNHJTZVKSWBUKDVPLSYS3Q2LX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PKNHJTZVKSWBUKDVPLSYS3Q2LX/action/storage_attestation","attest_author":"https://pith.science/pith/PKNHJTZVKSWBUKDVPLSYS3Q2LX/action/author_attestation","sign_citation":"https://pith.science/pith/PKNHJTZVKSWBUKDVPLSYS3Q2LX/action/citation_signature","submit_replication":"https://pith.science/pith/PKNHJTZVKSWBUKDVPLSYS3Q2LX/action/replication_record"}},"created_at":"2026-07-05T01:47:29.250014+00:00","updated_at":"2026-07-05T01:47:29.250014+00:00"}