{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:RG6K2SBIBFOYGJWKMD5QVT53GJ","short_pith_number":"pith:RG6K2SBI","schema_version":"1.0","canonical_sha256":"89bcad4828095d8326ca60fb0acfbb32678f92e2b02c274d07fb2b7ffce2aa21","source":{"kind":"arxiv","id":"2310.02984","version":2},"attestation_state":"computed","paper":{"title":"Scaling Laws for Associative Memories","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG","cs.NE"],"primary_cat":"stat.ML","authors_text":"Alberto Bietti, Elvis Dohmatob, Vivien Cabannes","submitted_at":"2023-10-04T17:20:34Z","abstract_excerpt":"Learning arguably involves the discovery and memorization of abstract rules. The aim of this paper is to study associative memory mechanisms. Our model is based on high-dimensional matrices consisting of outer products of embeddings, which relates to the inner layers of transformer language models. We derive precise scaling laws with respect to sample size and parameter size, and discuss the statistical efficiency of different estimators, including optimization-based algorithms. We provide extensive numerical experiments to validate and interpret theoretical results, including fine-grained vis"},"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":"2310.02984","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"stat.ML","submitted_at":"2023-10-04T17:20:34Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG","cs.NE"],"title_canon_sha256":"c0e04df4eb8f2df819efe4aaa352127911af69df21e02356dee74f94c1d1c180","abstract_canon_sha256":"f351968e1e7121429ef0ad57650d41e783d4d748e241fcc6e07f64d5e450e34b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:47:26.873892Z","signature_b64":"nR8tzHl3nitiYMVCbwZ7RyIpNIrJc7ENgjfbdjcoSM6ESTtuYrpk29c/9jRs5wuZd0C98XfX7zhwKqMxianoCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"89bcad4828095d8326ca60fb0acfbb32678f92e2b02c274d07fb2b7ffce2aa21","last_reissued_at":"2026-07-05T07:47:26.873313Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:47:26.873313Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Scaling Laws for Associative Memories","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG","cs.NE"],"primary_cat":"stat.ML","authors_text":"Alberto Bietti, Elvis Dohmatob, Vivien Cabannes","submitted_at":"2023-10-04T17:20:34Z","abstract_excerpt":"Learning arguably involves the discovery and memorization of abstract rules. The aim of this paper is to study associative memory mechanisms. Our model is based on high-dimensional matrices consisting of outer products of embeddings, which relates to the inner layers of transformer language models. We derive precise scaling laws with respect to sample size and parameter size, and discuss the statistical efficiency of different estimators, including optimization-based algorithms. We provide extensive numerical experiments to validate and interpret theoretical results, including fine-grained vis"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.02984","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/2310.02984/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":"2310.02984","created_at":"2026-07-05T07:47:26.873398+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.02984v2","created_at":"2026-07-05T07:47:26.873398+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.02984","created_at":"2026-07-05T07:47:26.873398+00:00"},{"alias_kind":"pith_short_12","alias_value":"RG6K2SBIBFOY","created_at":"2026-07-05T07:47:26.873398+00:00"},{"alias_kind":"pith_short_16","alias_value":"RG6K2SBIBFOYGJWK","created_at":"2026-07-05T07:47:26.873398+00:00"},{"alias_kind":"pith_short_8","alias_value":"RG6K2SBI","created_at":"2026-07-05T07:47:26.873398+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2508.00901","citing_title":"Provable Knowledge Acquisition and Extraction in One-Layer Transformers","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05189","citing_title":"Sharp Capacity Thresholds in Linear Associative Memory: From Winner-Take-All to Listwise Retrieval","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RG6K2SBIBFOYGJWKMD5QVT53GJ","json":"https://pith.science/pith/RG6K2SBIBFOYGJWKMD5QVT53GJ.json","graph_json":"https://pith.science/api/pith-number/RG6K2SBIBFOYGJWKMD5QVT53GJ/graph.json","events_json":"https://pith.science/api/pith-number/RG6K2SBIBFOYGJWKMD5QVT53GJ/events.json","paper":"https://pith.science/paper/RG6K2SBI"},"agent_actions":{"view_html":"https://pith.science/pith/RG6K2SBIBFOYGJWKMD5QVT53GJ","download_json":"https://pith.science/pith/RG6K2SBIBFOYGJWKMD5QVT53GJ.json","view_paper":"https://pith.science/paper/RG6K2SBI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.02984&json=true","fetch_graph":"https://pith.science/api/pith-number/RG6K2SBIBFOYGJWKMD5QVT53GJ/graph.json","fetch_events":"https://pith.science/api/pith-number/RG6K2SBIBFOYGJWKMD5QVT53GJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RG6K2SBIBFOYGJWKMD5QVT53GJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RG6K2SBIBFOYGJWKMD5QVT53GJ/action/storage_attestation","attest_author":"https://pith.science/pith/RG6K2SBIBFOYGJWKMD5QVT53GJ/action/author_attestation","sign_citation":"https://pith.science/pith/RG6K2SBIBFOYGJWKMD5QVT53GJ/action/citation_signature","submit_replication":"https://pith.science/pith/RG6K2SBIBFOYGJWKMD5QVT53GJ/action/replication_record"}},"created_at":"2026-07-05T07:47:26.873398+00:00","updated_at":"2026-07-05T07:47:26.873398+00:00"}