{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:IQKLUHI6AVZDNBKXY4JQX4P43X","short_pith_number":"pith:IQKLUHI6","schema_version":"1.0","canonical_sha256":"4414ba1d1e0572368557c7130bf1fcddef2607e9d2d3cd4203c22ff1dc86d879","source":{"kind":"arxiv","id":"2603.11021","version":2},"attestation_state":"computed","paper":{"title":"Leech Lattice Vector Quantization for Efficient LLM Compression","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Markus Nagel, Mart van Baalen, Paul Whatmough, Tycho F. A. van der Ouderaa","submitted_at":"2026-03-11T17:48:45Z","abstract_excerpt":"Scalar quantization of large language models (LLMs) is fundamentally limited by information-theoretic bounds. While vector quantization (VQ) overcomes these limits by encoding blocks of parameters jointly, practical implementations must avoid the need for expensive lookup mechanisms or other explicit codebook storage. Lattice approaches address this through highly structured and dense packing. This paper explores the Leech lattice, which, with its optimal sphere packing and kissing configurations at 24 dimensions, is the highest dimensional lattice known with such optimal properties. To make t"},"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":"2603.11021","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-03-11T17:48:45Z","cross_cats_sorted":[],"title_canon_sha256":"97d5654ded07a38517470660c20aca9ec5c016173781a3e0f591d2d2b389b765","abstract_canon_sha256":"b0bda2db9026333261b899a3803cd09c7572fce4e037ba73e9cfb956bb9635c0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-08T01:19:11.465991Z","signature_b64":"OQKfDIq6QC9fU9HNupS1FGdWBEyWg4C8WXfFs9KW50CoXYWeV/VJaW2gcnyfKLAnAo6BoDm+4Q85x6xYWi8pAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4414ba1d1e0572368557c7130bf1fcddef2607e9d2d3cd4203c22ff1dc86d879","last_reissued_at":"2026-07-08T01:19:11.465427Z","signature_status":"signed_v1","first_computed_at":"2026-07-08T01:19:11.465427Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Leech Lattice Vector Quantization for Efficient LLM Compression","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Markus Nagel, Mart van Baalen, Paul Whatmough, Tycho F. A. van der Ouderaa","submitted_at":"2026-03-11T17:48:45Z","abstract_excerpt":"Scalar quantization of large language models (LLMs) is fundamentally limited by information-theoretic bounds. While vector quantization (VQ) overcomes these limits by encoding blocks of parameters jointly, practical implementations must avoid the need for expensive lookup mechanisms or other explicit codebook storage. Lattice approaches address this through highly structured and dense packing. This paper explores the Leech lattice, which, with its optimal sphere packing and kissing configurations at 24 dimensions, is the highest dimensional lattice known with such optimal properties. To make t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.11021","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/2603.11021/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":"2603.11021","created_at":"2026-07-08T01:19:11.465492+00:00"},{"alias_kind":"arxiv_version","alias_value":"2603.11021v2","created_at":"2026-07-08T01:19:11.465492+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.11021","created_at":"2026-07-08T01:19:11.465492+00:00"},{"alias_kind":"pith_short_12","alias_value":"IQKLUHI6AVZD","created_at":"2026-07-08T01:19:11.465492+00:00"},{"alias_kind":"pith_short_16","alias_value":"IQKLUHI6AVZDNBKX","created_at":"2026-07-08T01:19:11.465492+00:00"},{"alias_kind":"pith_short_8","alias_value":"IQKLUHI6","created_at":"2026-07-08T01:19:11.465492+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2607.08643","citing_title":"BiSCo-LLM: Lookup-Free Binary Spherical Coding for Extreme Low-Bit Large Language Model Compression","ref_index":28,"is_internal_anchor":true},{"citing_arxiv_id":"2605.19972","citing_title":"Block-Sphere Vector Quantization","ref_index":29,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IQKLUHI6AVZDNBKXY4JQX4P43X","json":"https://pith.science/pith/IQKLUHI6AVZDNBKXY4JQX4P43X.json","graph_json":"https://pith.science/api/pith-number/IQKLUHI6AVZDNBKXY4JQX4P43X/graph.json","events_json":"https://pith.science/api/pith-number/IQKLUHI6AVZDNBKXY4JQX4P43X/events.json","paper":"https://pith.science/paper/IQKLUHI6"},"agent_actions":{"view_html":"https://pith.science/pith/IQKLUHI6AVZDNBKXY4JQX4P43X","download_json":"https://pith.science/pith/IQKLUHI6AVZDNBKXY4JQX4P43X.json","view_paper":"https://pith.science/paper/IQKLUHI6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2603.11021&json=true","fetch_graph":"https://pith.science/api/pith-number/IQKLUHI6AVZDNBKXY4JQX4P43X/graph.json","fetch_events":"https://pith.science/api/pith-number/IQKLUHI6AVZDNBKXY4JQX4P43X/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IQKLUHI6AVZDNBKXY4JQX4P43X/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IQKLUHI6AVZDNBKXY4JQX4P43X/action/storage_attestation","attest_author":"https://pith.science/pith/IQKLUHI6AVZDNBKXY4JQX4P43X/action/author_attestation","sign_citation":"https://pith.science/pith/IQKLUHI6AVZDNBKXY4JQX4P43X/action/citation_signature","submit_replication":"https://pith.science/pith/IQKLUHI6AVZDNBKXY4JQX4P43X/action/replication_record"}},"created_at":"2026-07-08T01:19:11.465492+00:00","updated_at":"2026-07-08T01:19:11.465492+00:00"}