{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:OS7YGXMA4XNSS6OPDAITKU66Z7","short_pith_number":"pith:OS7YGXMA","schema_version":"1.0","canonical_sha256":"74bf835d80e5db2979cf18113553decfd39c9ff98a14eabb6c3cd34b061563ab","source":{"kind":"arxiv","id":"2404.03827","version":3},"attestation_state":"computed","paper":{"title":"Uniform Memory Retrieval with Larger Capacity for Modern Hopfield Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Dennis Wu, Han Liu, Jerry Yao-Chieh Hu, Teng-Yun Hsiao","submitted_at":"2024-04-04T23:05:30Z","abstract_excerpt":"We propose a two-stage memory retrieval dynamics for modern Hopfield models, termed $\\mathtt{U\\text{-}Hop}$, with enhanced memory capacity. Our key contribution is a learnable feature map $\\Phi$ which transforms the Hopfield energy function into kernel space. This transformation ensures convergence between the local minima of energy and the fixed points of retrieval dynamics within the kernel space. Consequently, the kernel norm induced by $\\Phi$ serves as a novel similarity measure. It utilizes the stored memory patterns as learning data to enhance memory capacity across all modern Hopfield m"},"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":"2404.03827","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-04-04T23:05:30Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"c3ec91df38c407f5d585df5c3e6b56af841f45878ca328b0198a7edc178ec2bb","abstract_canon_sha256":"41b43bf28e5d1164119331b39ec3b782aa063e86556afc65b42190c55f008110"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:33:30.586835Z","signature_b64":"7Mw2YZOCEYc+ypTyaBG2DQLOFAslC+B53G9vy2Or5ABO5f9COyUBXJiqFmXS0iY9MYWdp5uDF9F6Am3wQJhjCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"74bf835d80e5db2979cf18113553decfd39c9ff98a14eabb6c3cd34b061563ab","last_reissued_at":"2026-07-05T09:33:30.586358Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:33:30.586358Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Uniform Memory Retrieval with Larger Capacity for Modern Hopfield Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Dennis Wu, Han Liu, Jerry Yao-Chieh Hu, Teng-Yun Hsiao","submitted_at":"2024-04-04T23:05:30Z","abstract_excerpt":"We propose a two-stage memory retrieval dynamics for modern Hopfield models, termed $\\mathtt{U\\text{-}Hop}$, with enhanced memory capacity. Our key contribution is a learnable feature map $\\Phi$ which transforms the Hopfield energy function into kernel space. This transformation ensures convergence between the local minima of energy and the fixed points of retrieval dynamics within the kernel space. Consequently, the kernel norm induced by $\\Phi$ serves as a novel similarity measure. It utilizes the stored memory patterns as learning data to enhance memory capacity across all modern Hopfield m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.03827","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/2404.03827/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":"2404.03827","created_at":"2026-07-05T09:33:30.586420+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.03827v3","created_at":"2026-07-05T09:33:30.586420+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.03827","created_at":"2026-07-05T09:33:30.586420+00:00"},{"alias_kind":"pith_short_12","alias_value":"OS7YGXMA4XNS","created_at":"2026-07-05T09:33:30.586420+00:00"},{"alias_kind":"pith_short_16","alias_value":"OS7YGXMA4XNSS6OP","created_at":"2026-07-05T09:33:30.586420+00:00"},{"alias_kind":"pith_short_8","alias_value":"OS7YGXMA","created_at":"2026-07-05T09:33:30.586420+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.11391","citing_title":"Recursive Binding on a Budget: Subspace Carving in Order-p Tensor Memories","ref_index":39,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OS7YGXMA4XNSS6OPDAITKU66Z7","json":"https://pith.science/pith/OS7YGXMA4XNSS6OPDAITKU66Z7.json","graph_json":"https://pith.science/api/pith-number/OS7YGXMA4XNSS6OPDAITKU66Z7/graph.json","events_json":"https://pith.science/api/pith-number/OS7YGXMA4XNSS6OPDAITKU66Z7/events.json","paper":"https://pith.science/paper/OS7YGXMA"},"agent_actions":{"view_html":"https://pith.science/pith/OS7YGXMA4XNSS6OPDAITKU66Z7","download_json":"https://pith.science/pith/OS7YGXMA4XNSS6OPDAITKU66Z7.json","view_paper":"https://pith.science/paper/OS7YGXMA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.03827&json=true","fetch_graph":"https://pith.science/api/pith-number/OS7YGXMA4XNSS6OPDAITKU66Z7/graph.json","fetch_events":"https://pith.science/api/pith-number/OS7YGXMA4XNSS6OPDAITKU66Z7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OS7YGXMA4XNSS6OPDAITKU66Z7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OS7YGXMA4XNSS6OPDAITKU66Z7/action/storage_attestation","attest_author":"https://pith.science/pith/OS7YGXMA4XNSS6OPDAITKU66Z7/action/author_attestation","sign_citation":"https://pith.science/pith/OS7YGXMA4XNSS6OPDAITKU66Z7/action/citation_signature","submit_replication":"https://pith.science/pith/OS7YGXMA4XNSS6OPDAITKU66Z7/action/replication_record"}},"created_at":"2026-07-05T09:33:30.586420+00:00","updated_at":"2026-07-05T09:33:30.586420+00:00"}