{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:LHIFZPQB6CSQZPW3FJPZLIJ7CF","short_pith_number":"pith:LHIFZPQB","schema_version":"1.0","canonical_sha256":"59d05cbe01f0a50cbedb2a5f95a13f1177920c83ad5fa1390c4a9cadbc36d169","source":{"kind":"arxiv","id":"2502.05164","version":2},"attestation_state":"computed","paper":{"title":"In-context denoising with one-layer transformers: connections between attention and associative memory retrieval","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cond-mat.dis-nn"],"primary_cat":"cs.LG","authors_text":"Alberto Bietti, Anirvan M. Sengupta, Matthew Smart","submitted_at":"2025-02-07T18:48:25Z","abstract_excerpt":"We introduce in-context denoising, a task that refines the connection between attention-based architectures and dense associative memory (DAM) networks, also known as modern Hopfield networks. Using a Bayesian framework, we show theoretically and empirically that certain restricted denoising problems can be solved optimally even by a single-layer transformer. We demonstrate that a trained attention layer processes each denoising prompt by performing a single gradient descent update on a context-aware DAM energy landscape, where context tokens serve as associative memories and the query token a"},"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":"2502.05164","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-07T18:48:25Z","cross_cats_sorted":["cond-mat.dis-nn"],"title_canon_sha256":"aeed6b594b0f8801efe161dbbc45c9202c5183d9079f8de104a1ba9f9d3407e2","abstract_canon_sha256":"e22ecffce09592dbd5b79769ec2e38890f6d7685657860ed564c7e31fefe1457"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:16:54.779528Z","signature_b64":"HbJSLfzL/kldROavfneC2FYPZipdB1IYwWyod4WY7yeWJ7u2W04/ZQDTSGbZuGlQH4YvlFoertWdlEtBc6hICg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"59d05cbe01f0a50cbedb2a5f95a13f1177920c83ad5fa1390c4a9cadbc36d169","last_reissued_at":"2026-07-05T11:16:54.779007Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:16:54.779007Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"In-context denoising with one-layer transformers: connections between attention and associative memory retrieval","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cond-mat.dis-nn"],"primary_cat":"cs.LG","authors_text":"Alberto Bietti, Anirvan M. Sengupta, Matthew Smart","submitted_at":"2025-02-07T18:48:25Z","abstract_excerpt":"We introduce in-context denoising, a task that refines the connection between attention-based architectures and dense associative memory (DAM) networks, also known as modern Hopfield networks. Using a Bayesian framework, we show theoretically and empirically that certain restricted denoising problems can be solved optimally even by a single-layer transformer. We demonstrate that a trained attention layer processes each denoising prompt by performing a single gradient descent update on a context-aware DAM energy landscape, where context tokens serve as associative memories and the query token a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.05164","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/2502.05164/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":"2502.05164","created_at":"2026-07-05T11:16:54.779064+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.05164v2","created_at":"2026-07-05T11:16:54.779064+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.05164","created_at":"2026-07-05T11:16:54.779064+00:00"},{"alias_kind":"pith_short_12","alias_value":"LHIFZPQB6CSQ","created_at":"2026-07-05T11:16:54.779064+00:00"},{"alias_kind":"pith_short_16","alias_value":"LHIFZPQB6CSQZPW3","created_at":"2026-07-05T11:16:54.779064+00:00"},{"alias_kind":"pith_short_8","alias_value":"LHIFZPQB","created_at":"2026-07-05T11:16:54.779064+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/LHIFZPQB6CSQZPW3FJPZLIJ7CF","json":"https://pith.science/pith/LHIFZPQB6CSQZPW3FJPZLIJ7CF.json","graph_json":"https://pith.science/api/pith-number/LHIFZPQB6CSQZPW3FJPZLIJ7CF/graph.json","events_json":"https://pith.science/api/pith-number/LHIFZPQB6CSQZPW3FJPZLIJ7CF/events.json","paper":"https://pith.science/paper/LHIFZPQB"},"agent_actions":{"view_html":"https://pith.science/pith/LHIFZPQB6CSQZPW3FJPZLIJ7CF","download_json":"https://pith.science/pith/LHIFZPQB6CSQZPW3FJPZLIJ7CF.json","view_paper":"https://pith.science/paper/LHIFZPQB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.05164&json=true","fetch_graph":"https://pith.science/api/pith-number/LHIFZPQB6CSQZPW3FJPZLIJ7CF/graph.json","fetch_events":"https://pith.science/api/pith-number/LHIFZPQB6CSQZPW3FJPZLIJ7CF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LHIFZPQB6CSQZPW3FJPZLIJ7CF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LHIFZPQB6CSQZPW3FJPZLIJ7CF/action/storage_attestation","attest_author":"https://pith.science/pith/LHIFZPQB6CSQZPW3FJPZLIJ7CF/action/author_attestation","sign_citation":"https://pith.science/pith/LHIFZPQB6CSQZPW3FJPZLIJ7CF/action/citation_signature","submit_replication":"https://pith.science/pith/LHIFZPQB6CSQZPW3FJPZLIJ7CF/action/replication_record"}},"created_at":"2026-07-05T11:16:54.779064+00:00","updated_at":"2026-07-05T11:16:54.779064+00:00"}