{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:GN7LCM6CZOBYJ7GTNJAUPQNJL7","short_pith_number":"pith:GN7LCM6C","schema_version":"1.0","canonical_sha256":"337eb133c2cb8384fcd36a4147c1a95fc43191c1fa0594c14179cfe1f85a3c61","source":{"kind":"arxiv","id":"2410.01537","version":2},"attestation_state":"computed","paper":{"title":"Attention layers provably solve single-location regression","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Claire Boyer, G\\'erard Biau, Pierre Marion, Rapha\\\"el Berthier","submitted_at":"2024-10-02T13:28:02Z","abstract_excerpt":"Attention-based models, such as Transformer, excel across various tasks but lack a comprehensive theoretical understanding, especially regarding token-wise sparsity and internal linear representations. To address this gap, we introduce the single-location regression task, where only one token in a sequence determines the output, and its position is a latent random variable, retrievable via a linear projection of the input. To solve this task, we propose a dedicated predictor, which turns out to be a simplified version of a non-linear self-attention layer. We study its theoretical properties, b"},"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":"2410.01537","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2024-10-02T13:28:02Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"2f46bffa62fae48a7a9d23c9ba9db7ccc588eff60a40a605f0c630988af1ca12","abstract_canon_sha256":"b05316fcc0a0a1adf311f498563f546da915c347598e98d0a5dff7aebda1169f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:19:55.971001Z","signature_b64":"3gzX3YYh9RvPeB7z7hAyA/Oqbju1PkOV3GLw08npJCyRVhsL+c7JzevSdik1QjQItA/0Dgw8SrbEYZuuapwIBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"337eb133c2cb8384fcd36a4147c1a95fc43191c1fa0594c14179cfe1f85a3c61","last_reissued_at":"2026-07-05T10:19:55.970385Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:19:55.970385Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Attention layers provably solve single-location regression","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Claire Boyer, G\\'erard Biau, Pierre Marion, Rapha\\\"el Berthier","submitted_at":"2024-10-02T13:28:02Z","abstract_excerpt":"Attention-based models, such as Transformer, excel across various tasks but lack a comprehensive theoretical understanding, especially regarding token-wise sparsity and internal linear representations. To address this gap, we introduce the single-location regression task, where only one token in a sequence determines the output, and its position is a latent random variable, retrievable via a linear projection of the input. To solve this task, we propose a dedicated predictor, which turns out to be a simplified version of a non-linear self-attention layer. We study its theoretical properties, b"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.01537","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/2410.01537/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":"2410.01537","created_at":"2026-07-05T10:19:55.970488+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.01537v2","created_at":"2026-07-05T10:19:55.970488+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.01537","created_at":"2026-07-05T10:19:55.970488+00:00"},{"alias_kind":"pith_short_12","alias_value":"GN7LCM6CZOBY","created_at":"2026-07-05T10:19:55.970488+00:00"},{"alias_kind":"pith_short_16","alias_value":"GN7LCM6CZOBYJ7GT","created_at":"2026-07-05T10:19:55.970488+00:00"},{"alias_kind":"pith_short_8","alias_value":"GN7LCM6C","created_at":"2026-07-05T10:19:55.970488+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/GN7LCM6CZOBYJ7GTNJAUPQNJL7","json":"https://pith.science/pith/GN7LCM6CZOBYJ7GTNJAUPQNJL7.json","graph_json":"https://pith.science/api/pith-number/GN7LCM6CZOBYJ7GTNJAUPQNJL7/graph.json","events_json":"https://pith.science/api/pith-number/GN7LCM6CZOBYJ7GTNJAUPQNJL7/events.json","paper":"https://pith.science/paper/GN7LCM6C"},"agent_actions":{"view_html":"https://pith.science/pith/GN7LCM6CZOBYJ7GTNJAUPQNJL7","download_json":"https://pith.science/pith/GN7LCM6CZOBYJ7GTNJAUPQNJL7.json","view_paper":"https://pith.science/paper/GN7LCM6C","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.01537&json=true","fetch_graph":"https://pith.science/api/pith-number/GN7LCM6CZOBYJ7GTNJAUPQNJL7/graph.json","fetch_events":"https://pith.science/api/pith-number/GN7LCM6CZOBYJ7GTNJAUPQNJL7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GN7LCM6CZOBYJ7GTNJAUPQNJL7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GN7LCM6CZOBYJ7GTNJAUPQNJL7/action/storage_attestation","attest_author":"https://pith.science/pith/GN7LCM6CZOBYJ7GTNJAUPQNJL7/action/author_attestation","sign_citation":"https://pith.science/pith/GN7LCM6CZOBYJ7GTNJAUPQNJL7/action/citation_signature","submit_replication":"https://pith.science/pith/GN7LCM6CZOBYJ7GTNJAUPQNJL7/action/replication_record"}},"created_at":"2026-07-05T10:19:55.970488+00:00","updated_at":"2026-07-05T10:19:55.970488+00:00"}