{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:LKRKVOLLKXZLRDE6DIGJWWPRI5","short_pith_number":"pith:LKRKVOLL","schema_version":"1.0","canonical_sha256":"5aa2aab96b55f2b88c9e1a0c9b59f14775dadf1f3cf2aa78a6c780dbb04d46ea","source":{"kind":"arxiv","id":"2405.06394","version":3},"attestation_state":"computed","paper":{"title":"Memory Mosaics","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.NE"],"primary_cat":"cs.LG","authors_text":"Beidi Chen, Jianyu Zhang, L\\'eon Bottou, Niklas Nolte, Ranajoy Sadhukhan","submitted_at":"2024-05-10T11:08:20Z","abstract_excerpt":"Memory Mosaics are networks of associative memories working in concert to achieve a prediction task of interest. Like transformers, memory mosaics possess compositional capabilities and in-context learning capabilities. Unlike transformers, memory mosaics achieve these capabilities in comparatively transparent way (\"predictive disentanglement\"). We illustrate these capabilities on a toy example and also show that memory mosaics perform as well or better than transformers on medium-scale language modeling tasks."},"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":"2405.06394","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2024-05-10T11:08:20Z","cross_cats_sorted":["cs.AI","cs.NE"],"title_canon_sha256":"51e47b68cce60ff8b30ab783af08c5ad0ca1875ffb8427a2ec8e42c4f365c547","abstract_canon_sha256":"89f99722bf338ee1b3e344bf38b3e673654dca73489f3cb073549c9b67a96d9f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:21:19.091559Z","signature_b64":"iWAFpOJ6HEZRQWI1oziRoBt3zYnTDfO+mdC+Vso0CaxNvK+rDKVnjOqfBL/w2bKylju09lhGq05c8HsBMLQfBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5aa2aab96b55f2b88c9e1a0c9b59f14775dadf1f3cf2aa78a6c780dbb04d46ea","last_reissued_at":"2026-07-05T10:21:19.091068Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:21:19.091068Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Memory Mosaics","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.NE"],"primary_cat":"cs.LG","authors_text":"Beidi Chen, Jianyu Zhang, L\\'eon Bottou, Niklas Nolte, Ranajoy Sadhukhan","submitted_at":"2024-05-10T11:08:20Z","abstract_excerpt":"Memory Mosaics are networks of associative memories working in concert to achieve a prediction task of interest. Like transformers, memory mosaics possess compositional capabilities and in-context learning capabilities. Unlike transformers, memory mosaics achieve these capabilities in comparatively transparent way (\"predictive disentanglement\"). We illustrate these capabilities on a toy example and also show that memory mosaics perform as well or better than transformers on medium-scale language modeling tasks."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.06394","kind":"arxiv","version":3},"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/2405.06394/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":"2405.06394","created_at":"2026-07-05T10:21:19.091119+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.06394v3","created_at":"2026-07-05T10:21:19.091119+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.06394","created_at":"2026-07-05T10:21:19.091119+00:00"},{"alias_kind":"pith_short_12","alias_value":"LKRKVOLLKXZL","created_at":"2026-07-05T10:21:19.091119+00:00"},{"alias_kind":"pith_short_16","alias_value":"LKRKVOLLKXZLRDE6","created_at":"2026-07-05T10:21:19.091119+00:00"},{"alias_kind":"pith_short_8","alias_value":"LKRKVOLL","created_at":"2026-07-05T10:21:19.091119+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2501.00663","citing_title":"Titans: Learning to Memorize at Test Time","ref_index":133,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10537","citing_title":"Mela: Test-Time Memory Consolidation based on Transformation Hypothesis","ref_index":25,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LKRKVOLLKXZLRDE6DIGJWWPRI5","json":"https://pith.science/pith/LKRKVOLLKXZLRDE6DIGJWWPRI5.json","graph_json":"https://pith.science/api/pith-number/LKRKVOLLKXZLRDE6DIGJWWPRI5/graph.json","events_json":"https://pith.science/api/pith-number/LKRKVOLLKXZLRDE6DIGJWWPRI5/events.json","paper":"https://pith.science/paper/LKRKVOLL"},"agent_actions":{"view_html":"https://pith.science/pith/LKRKVOLLKXZLRDE6DIGJWWPRI5","download_json":"https://pith.science/pith/LKRKVOLLKXZLRDE6DIGJWWPRI5.json","view_paper":"https://pith.science/paper/LKRKVOLL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.06394&json=true","fetch_graph":"https://pith.science/api/pith-number/LKRKVOLLKXZLRDE6DIGJWWPRI5/graph.json","fetch_events":"https://pith.science/api/pith-number/LKRKVOLLKXZLRDE6DIGJWWPRI5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LKRKVOLLKXZLRDE6DIGJWWPRI5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LKRKVOLLKXZLRDE6DIGJWWPRI5/action/storage_attestation","attest_author":"https://pith.science/pith/LKRKVOLLKXZLRDE6DIGJWWPRI5/action/author_attestation","sign_citation":"https://pith.science/pith/LKRKVOLLKXZLRDE6DIGJWWPRI5/action/citation_signature","submit_replication":"https://pith.science/pith/LKRKVOLLKXZLRDE6DIGJWWPRI5/action/replication_record"}},"created_at":"2026-07-05T10:21:19.091119+00:00","updated_at":"2026-07-05T10:21:19.091119+00:00"}