{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:DVA3XILXLPMF5ZNITGCL3GT3OJ","short_pith_number":"pith:DVA3XILX","schema_version":"1.0","canonical_sha256":"1d41bba1775bd85ee5a89984bd9a7b726b439cfd1c00b3dcf649afb93e1a74b8","source":{"kind":"arxiv","id":"2502.09720","version":3},"attestation_state":"computed","paper":{"title":"NestQuant: Nested Lattice Quantization for Matrix Products and LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.IT","math.IT"],"primary_cat":"cs.LG","authors_text":"Eitan Porat, Or Ordentlich, Semyon Savkin, Yury Polyanskiy","submitted_at":"2025-02-13T19:11:40Z","abstract_excerpt":"Post-training quantization (PTQ) has emerged as a critical technique for efficient deployment of large language models (LLMs). This work proposes NestQuant, a novel PTQ scheme for weights and activations that is based on self-similar nested lattices. Recent works have mathematically shown such quantizers to be information-theoretically optimal for low-precision matrix multiplication. We implement a practical low-complexity version of NestQuant based on Gosset lattice, making it a drop-in quantizer for any matrix multiplication step (e.g., in self-attention, MLP etc). For example, NestQuant qua"},"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":"2502.09720","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-13T19:11:40Z","cross_cats_sorted":["cs.AI","cs.IT","math.IT"],"title_canon_sha256":"e2c049a9826cc8e879fd346b8eedfc59970452fbf540ebd5ecf32abbfac599f7","abstract_canon_sha256":"2596bff04de777165fb4931e051a2cf6f716ac5428d6cac30700cfa76cddcea6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:43:30.710586Z","signature_b64":"yvohkqS8/jAylFcS3UzEeQDlITzqVrddhkcPBjV6ExNLFJ4rO96hcWmGYgFdSooCCm6raU4Ty1nfAvqYuVlMDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1d41bba1775bd85ee5a89984bd9a7b726b439cfd1c00b3dcf649afb93e1a74b8","last_reissued_at":"2026-07-05T11:43:30.710057Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:43:30.710057Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"NestQuant: Nested Lattice Quantization for Matrix Products and LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.IT","math.IT"],"primary_cat":"cs.LG","authors_text":"Eitan Porat, Or Ordentlich, Semyon Savkin, Yury Polyanskiy","submitted_at":"2025-02-13T19:11:40Z","abstract_excerpt":"Post-training quantization (PTQ) has emerged as a critical technique for efficient deployment of large language models (LLMs). This work proposes NestQuant, a novel PTQ scheme for weights and activations that is based on self-similar nested lattices. Recent works have mathematically shown such quantizers to be information-theoretically optimal for low-precision matrix multiplication. We implement a practical low-complexity version of NestQuant based on Gosset lattice, making it a drop-in quantizer for any matrix multiplication step (e.g., in self-attention, MLP etc). For example, NestQuant qua"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.09720","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/2502.09720/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":"2502.09720","created_at":"2026-07-05T11:43:30.710119+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.09720v3","created_at":"2026-07-05T11:43:30.710119+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.09720","created_at":"2026-07-05T11:43:30.710119+00:00"},{"alias_kind":"pith_short_12","alias_value":"DVA3XILXLPMF","created_at":"2026-07-05T11:43:30.710119+00:00"},{"alias_kind":"pith_short_16","alias_value":"DVA3XILXLPMF5ZNI","created_at":"2026-07-05T11:43:30.710119+00:00"},{"alias_kind":"pith_short_8","alias_value":"DVA3XILX","created_at":"2026-07-05T11:43:30.710119+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.23406","citing_title":"HyperQuant: A Rate-Distortion-Optimal Quantization Pipeline for Large Language and Diffusion Models","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13768","citing_title":"High-Rate Quantized Matrix Multiplication II","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2601.17187","citing_title":"High-Rate Quantized Matrix Multiplication I","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13768","citing_title":"High-Rate Quantized Matrix Multiplication II","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DVA3XILXLPMF5ZNITGCL3GT3OJ","json":"https://pith.science/pith/DVA3XILXLPMF5ZNITGCL3GT3OJ.json","graph_json":"https://pith.science/api/pith-number/DVA3XILXLPMF5ZNITGCL3GT3OJ/graph.json","events_json":"https://pith.science/api/pith-number/DVA3XILXLPMF5ZNITGCL3GT3OJ/events.json","paper":"https://pith.science/paper/DVA3XILX"},"agent_actions":{"view_html":"https://pith.science/pith/DVA3XILXLPMF5ZNITGCL3GT3OJ","download_json":"https://pith.science/pith/DVA3XILXLPMF5ZNITGCL3GT3OJ.json","view_paper":"https://pith.science/paper/DVA3XILX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.09720&json=true","fetch_graph":"https://pith.science/api/pith-number/DVA3XILXLPMF5ZNITGCL3GT3OJ/graph.json","fetch_events":"https://pith.science/api/pith-number/DVA3XILXLPMF5ZNITGCL3GT3OJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DVA3XILXLPMF5ZNITGCL3GT3OJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DVA3XILXLPMF5ZNITGCL3GT3OJ/action/storage_attestation","attest_author":"https://pith.science/pith/DVA3XILXLPMF5ZNITGCL3GT3OJ/action/author_attestation","sign_citation":"https://pith.science/pith/DVA3XILXLPMF5ZNITGCL3GT3OJ/action/citation_signature","submit_replication":"https://pith.science/pith/DVA3XILXLPMF5ZNITGCL3GT3OJ/action/replication_record"}},"created_at":"2026-07-05T11:43:30.710119+00:00","updated_at":"2026-07-05T11:43:30.710119+00:00"}