{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:QNWJDFSKCM7SXZFKC3NOPSH5BL","short_pith_number":"pith:QNWJDFSK","schema_version":"1.0","canonical_sha256":"836c91964a133f2be4aa16dae7c8fd0af531af7f060f6a0141e8f9db7eed5402","source":{"kind":"arxiv","id":"2602.06694","version":2},"attestation_state":"computed","paper":{"title":"NanoQuant: Efficient Sub-1-Bit Quantization of Large Language Models","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Changdong Kim, Dongkyu Kim, Hyochan Chong, Minseop Choi","submitted_at":"2026-02-06T13:26:44Z","abstract_excerpt":"Weight-only quantization has become a standard approach for efficiently serving large language models (LLMs). However, existing methods fail to efficiently compress models to binary (1-bit) levels, as they either require large amounts of data and compute or incur additional storage. In this work, we propose NanoQuant, the first post-training quantization (PTQ) method to compress LLMs to both binary and sub-1-bit levels. NanoQuant formulates quantization as a low-rank binary factorization problem, and compresses full-precision weights to low-rank binary matrices and scales. Specifically, it uti"},"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":"2602.06694","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2026-02-06T13:26:44Z","cross_cats_sorted":[],"title_canon_sha256":"32cee76de8743bda061292ed785db40127ddfddc2c0c9234cd20b7f2c740ee8a","abstract_canon_sha256":"c157d0a68cf51282ba63a6d910f916fd91f4af8d8687feeee8e6c79c2efa8b60"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:04:24.731635Z","signature_b64":"Dz2uAE87DNq6ParvX5J5K0ijZiNvUsWtnuC6lDwrk/QzXes9f0eZ+/XQZiMWwQ7H0/rnblGRxP442wezNPZKCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"836c91964a133f2be4aa16dae7c8fd0af531af7f060f6a0141e8f9db7eed5402","last_reissued_at":"2026-05-20T00:04:24.730564Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:04:24.730564Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"NanoQuant: Efficient Sub-1-Bit Quantization of Large Language Models","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Changdong Kim, Dongkyu Kim, Hyochan Chong, Minseop Choi","submitted_at":"2026-02-06T13:26:44Z","abstract_excerpt":"Weight-only quantization has become a standard approach for efficiently serving large language models (LLMs). However, existing methods fail to efficiently compress models to binary (1-bit) levels, as they either require large amounts of data and compute or incur additional storage. In this work, we propose NanoQuant, the first post-training quantization (PTQ) method to compress LLMs to both binary and sub-1-bit levels. NanoQuant formulates quantization as a low-rank binary factorization problem, and compresses full-precision weights to low-rank binary matrices and scales. Specifically, it uti"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.06694","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/2602.06694/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":"2602.06694","created_at":"2026-05-20T00:04:24.730777+00:00"},{"alias_kind":"arxiv_version","alias_value":"2602.06694v2","created_at":"2026-05-20T00:04:24.730777+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.06694","created_at":"2026-05-20T00:04:24.730777+00:00"},{"alias_kind":"pith_short_12","alias_value":"QNWJDFSKCM7S","created_at":"2026-05-20T00:04:24.730777+00:00"},{"alias_kind":"pith_short_16","alias_value":"QNWJDFSKCM7SXZFK","created_at":"2026-05-20T00:04:24.730777+00:00"},{"alias_kind":"pith_short_8","alias_value":"QNWJDFSK","created_at":"2026-05-20T00:04:24.730777+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QNWJDFSKCM7SXZFKC3NOPSH5BL","json":"https://pith.science/pith/QNWJDFSKCM7SXZFKC3NOPSH5BL.json","graph_json":"https://pith.science/api/pith-number/QNWJDFSKCM7SXZFKC3NOPSH5BL/graph.json","events_json":"https://pith.science/api/pith-number/QNWJDFSKCM7SXZFKC3NOPSH5BL/events.json","paper":"https://pith.science/paper/QNWJDFSK"},"agent_actions":{"view_html":"https://pith.science/pith/QNWJDFSKCM7SXZFKC3NOPSH5BL","download_json":"https://pith.science/pith/QNWJDFSKCM7SXZFKC3NOPSH5BL.json","view_paper":"https://pith.science/paper/QNWJDFSK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2602.06694&json=true","fetch_graph":"https://pith.science/api/pith-number/QNWJDFSKCM7SXZFKC3NOPSH5BL/graph.json","fetch_events":"https://pith.science/api/pith-number/QNWJDFSKCM7SXZFKC3NOPSH5BL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QNWJDFSKCM7SXZFKC3NOPSH5BL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QNWJDFSKCM7SXZFKC3NOPSH5BL/action/storage_attestation","attest_author":"https://pith.science/pith/QNWJDFSKCM7SXZFKC3NOPSH5BL/action/author_attestation","sign_citation":"https://pith.science/pith/QNWJDFSKCM7SXZFKC3NOPSH5BL/action/citation_signature","submit_replication":"https://pith.science/pith/QNWJDFSKCM7SXZFKC3NOPSH5BL/action/replication_record"}},"created_at":"2026-05-20T00:04:24.730777+00:00","updated_at":"2026-05-20T00:04:24.730777+00:00"}