{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:ZHXF744I3RXGIKA57DAGK6DM2U","short_pith_number":"pith:ZHXF744I","schema_version":"1.0","canonical_sha256":"c9ee5ff388dc6e64281df8c065786cd525f96b8ebb1baf1a8ec0fd7060c13cc8","source":{"kind":"arxiv","id":"2410.18541","version":1},"attestation_state":"computed","paper":{"title":"On Explaining with Attention Matrices","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Nicholas Asher, Omar Naim","submitted_at":"2024-10-24T08:43:33Z","abstract_excerpt":"This paper explores the much discussed, possible explanatory link between attention weights (AW) in transformer models and predicted output. Contrary to intuition and early research on attention, more recent prior research has provided formal arguments and empirical evidence that AW are not explanatorily relevant. We show that the formal arguments are incorrect. We introduce and effectively compute efficient attention, which isolates the effective components of attention matrices in tasks and models in which AW play an explanatory role. We show that efficient attention has a causal role (provi"},"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":"2410.18541","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-24T08:43:33Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"494ec9cd1d0fda06f42ffc8a3c4173295c2ed9d95aaeeccb2fb5be01db7172e5","abstract_canon_sha256":"401d8686622f0be267103c53fde36ffde7bbe2768f73325d3d88b7081677cdd7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:25:10.017843Z","signature_b64":"4QiBXWGTDfwH1YS5fI74lJbsffHeA49/kGGRU3+u5r3eYsisO86B7ouwUz681eBCCsbv5XrkLeUitxrkCXfHCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c9ee5ff388dc6e64281df8c065786cd525f96b8ebb1baf1a8ec0fd7060c13cc8","last_reissued_at":"2026-07-05T09:25:10.017150Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:25:10.017150Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On Explaining with Attention Matrices","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Nicholas Asher, Omar Naim","submitted_at":"2024-10-24T08:43:33Z","abstract_excerpt":"This paper explores the much discussed, possible explanatory link between attention weights (AW) in transformer models and predicted output. Contrary to intuition and early research on attention, more recent prior research has provided formal arguments and empirical evidence that AW are not explanatorily relevant. We show that the formal arguments are incorrect. We introduce and effectively compute efficient attention, which isolates the effective components of attention matrices in tasks and models in which AW play an explanatory role. We show that efficient attention has a causal role (provi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.18541","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/2410.18541/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":"2410.18541","created_at":"2026-07-05T09:25:10.017217+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.18541v1","created_at":"2026-07-05T09:25:10.017217+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.18541","created_at":"2026-07-05T09:25:10.017217+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZHXF744I3RXG","created_at":"2026-07-05T09:25:10.017217+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZHXF744I3RXGIKA5","created_at":"2026-07-05T09:25:10.017217+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZHXF744I","created_at":"2026-07-05T09:25:10.017217+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/ZHXF744I3RXGIKA57DAGK6DM2U","json":"https://pith.science/pith/ZHXF744I3RXGIKA57DAGK6DM2U.json","graph_json":"https://pith.science/api/pith-number/ZHXF744I3RXGIKA57DAGK6DM2U/graph.json","events_json":"https://pith.science/api/pith-number/ZHXF744I3RXGIKA57DAGK6DM2U/events.json","paper":"https://pith.science/paper/ZHXF744I"},"agent_actions":{"view_html":"https://pith.science/pith/ZHXF744I3RXGIKA57DAGK6DM2U","download_json":"https://pith.science/pith/ZHXF744I3RXGIKA57DAGK6DM2U.json","view_paper":"https://pith.science/paper/ZHXF744I","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.18541&json=true","fetch_graph":"https://pith.science/api/pith-number/ZHXF744I3RXGIKA57DAGK6DM2U/graph.json","fetch_events":"https://pith.science/api/pith-number/ZHXF744I3RXGIKA57DAGK6DM2U/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZHXF744I3RXGIKA57DAGK6DM2U/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZHXF744I3RXGIKA57DAGK6DM2U/action/storage_attestation","attest_author":"https://pith.science/pith/ZHXF744I3RXGIKA57DAGK6DM2U/action/author_attestation","sign_citation":"https://pith.science/pith/ZHXF744I3RXGIKA57DAGK6DM2U/action/citation_signature","submit_replication":"https://pith.science/pith/ZHXF744I3RXGIKA57DAGK6DM2U/action/replication_record"}},"created_at":"2026-07-05T09:25:10.017217+00:00","updated_at":"2026-07-05T09:25:10.017217+00:00"}