{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:PLKQLDV54GXULAWHCDEOBQABO7","short_pith_number":"pith:PLKQLDV5","schema_version":"1.0","canonical_sha256":"7ad5058ebde1af4582c710c8e0c00177c8e99bab77dbacda4f17db4b39a739be","source":{"kind":"arxiv","id":"2310.18805","version":2},"attestation_state":"computed","paper":{"title":"Inverse distance weighting attention","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Calvin McCarter","submitted_at":"2023-10-28T20:11:01Z","abstract_excerpt":"We report the effects of replacing the scaled dot-product (within softmax) attention with the negative-log of Euclidean distance. This form of attention simplifies to inverse distance weighting interpolation. Used in simple one hidden layer networks and trained with vanilla cross-entropy loss on classification problems, it tends to produce a key matrix containing prototypes and a value matrix with corresponding logits. We also show that the resulting interpretable networks can be augmented with manually-constructed prototypes to perform low-impact handling of special cases."},"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":"2310.18805","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2023-10-28T20:11:01Z","cross_cats_sorted":[],"title_canon_sha256":"dc05f8e201a77aada0bd124529dbadcadb87e4a7732387c942b5993d6bd54143","abstract_canon_sha256":"1bb11c6ac65320cabc7bfd52afd681ab4006d94ed6cf4499a722b0c115895556"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:21:12.453657Z","signature_b64":"/B69eXT0alUWZSkQjdSKy9/B1USJBdoFHqc7fFBKgN2KhKTyVH+568WZVB6CoSbmzAk5Vnfe7zloOndrZ5VaBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7ad5058ebde1af4582c710c8e0c00177c8e99bab77dbacda4f17db4b39a739be","last_reissued_at":"2026-07-05T07:21:12.453145Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:21:12.453145Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Inverse distance weighting attention","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Calvin McCarter","submitted_at":"2023-10-28T20:11:01Z","abstract_excerpt":"We report the effects of replacing the scaled dot-product (within softmax) attention with the negative-log of Euclidean distance. This form of attention simplifies to inverse distance weighting interpolation. Used in simple one hidden layer networks and trained with vanilla cross-entropy loss on classification problems, it tends to produce a key matrix containing prototypes and a value matrix with corresponding logits. We also show that the resulting interpretable networks can be augmented with manually-constructed prototypes to perform low-impact handling of special cases."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.18805","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/2310.18805/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":"2310.18805","created_at":"2026-07-05T07:21:12.453220+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.18805v2","created_at":"2026-07-05T07:21:12.453220+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.18805","created_at":"2026-07-05T07:21:12.453220+00:00"},{"alias_kind":"pith_short_12","alias_value":"PLKQLDV54GXU","created_at":"2026-07-05T07:21:12.453220+00:00"},{"alias_kind":"pith_short_16","alias_value":"PLKQLDV54GXULAWH","created_at":"2026-07-05T07:21:12.453220+00:00"},{"alias_kind":"pith_short_8","alias_value":"PLKQLDV5","created_at":"2026-07-05T07:21:12.453220+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.05258","citing_title":"Harnessing Source Heterogeneity for Cluster-Structured Transfer Learning","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2509.04154","citing_title":"Robust Filter Attention: Self-Attention as Precision-Weighted State Estimation","ref_index":57,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PLKQLDV54GXULAWHCDEOBQABO7","json":"https://pith.science/pith/PLKQLDV54GXULAWHCDEOBQABO7.json","graph_json":"https://pith.science/api/pith-number/PLKQLDV54GXULAWHCDEOBQABO7/graph.json","events_json":"https://pith.science/api/pith-number/PLKQLDV54GXULAWHCDEOBQABO7/events.json","paper":"https://pith.science/paper/PLKQLDV5"},"agent_actions":{"view_html":"https://pith.science/pith/PLKQLDV54GXULAWHCDEOBQABO7","download_json":"https://pith.science/pith/PLKQLDV54GXULAWHCDEOBQABO7.json","view_paper":"https://pith.science/paper/PLKQLDV5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.18805&json=true","fetch_graph":"https://pith.science/api/pith-number/PLKQLDV54GXULAWHCDEOBQABO7/graph.json","fetch_events":"https://pith.science/api/pith-number/PLKQLDV54GXULAWHCDEOBQABO7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PLKQLDV54GXULAWHCDEOBQABO7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PLKQLDV54GXULAWHCDEOBQABO7/action/storage_attestation","attest_author":"https://pith.science/pith/PLKQLDV54GXULAWHCDEOBQABO7/action/author_attestation","sign_citation":"https://pith.science/pith/PLKQLDV54GXULAWHCDEOBQABO7/action/citation_signature","submit_replication":"https://pith.science/pith/PLKQLDV54GXULAWHCDEOBQABO7/action/replication_record"}},"created_at":"2026-07-05T07:21:12.453220+00:00","updated_at":"2026-07-05T07:21:12.453220+00:00"}