{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:AHPKUNJ7TPD4BKWNK26ZJHIG3T","short_pith_number":"pith:AHPKUNJ7","schema_version":"1.0","canonical_sha256":"01deaa353f9bc7c0aacd56bd949d06dce49a242fbf821d2bd10d7b04d2a6a5cf","source":{"kind":"arxiv","id":"2402.12065","version":2},"attestation_state":"computed","paper":{"title":"WKVQuant: Quantizing Weight and Key/Value Cache for Large Language Models Gains More","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Haojie Duanmu, Jianlong Wu, Liqiang Nie, Sifan Zhou, Yuxuan Yue, Zhihang Yuan","submitted_at":"2024-02-19T11:33:21Z","abstract_excerpt":"Large Language Models (LLMs) face significant deployment challenges due to their substantial memory requirements and the computational demands of auto-regressive text generation process. This paper addresses these challenges by focusing on the quantization of LLMs, a technique that reduces memory consumption by converting model parameters and activations into low-bit integers. We critically analyze the existing quantization approaches, identifying their limitations in balancing the accuracy and efficiency of the quantized LLMs. To advance beyond these limitations, we propose WKVQuant, a PTQ fr"},"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":"2402.12065","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-19T11:33:21Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"caf489465212f4b6d225725c5a3dce0012639fa73fb304ccb4db604a4ee64913","abstract_canon_sha256":"82773ad59a566819755f522afdab0d1e649f5cf026c8e10ac5063847d6b50876"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:47:10.347857Z","signature_b64":"krPJdYzTBYZ63phwI3XtKIDGJ6cVfJBwyx+mBHK8SUqkrB2UTAdBAosCf7314zICYoWpTyKevZ6VgIkiChvQDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"01deaa353f9bc7c0aacd56bd949d06dce49a242fbf821d2bd10d7b04d2a6a5cf","last_reissued_at":"2026-07-05T07:47:10.347377Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:47:10.347377Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"WKVQuant: Quantizing Weight and Key/Value Cache for Large Language Models Gains More","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Haojie Duanmu, Jianlong Wu, Liqiang Nie, Sifan Zhou, Yuxuan Yue, Zhihang Yuan","submitted_at":"2024-02-19T11:33:21Z","abstract_excerpt":"Large Language Models (LLMs) face significant deployment challenges due to their substantial memory requirements and the computational demands of auto-regressive text generation process. This paper addresses these challenges by focusing on the quantization of LLMs, a technique that reduces memory consumption by converting model parameters and activations into low-bit integers. We critically analyze the existing quantization approaches, identifying their limitations in balancing the accuracy and efficiency of the quantized LLMs. To advance beyond these limitations, we propose WKVQuant, a PTQ fr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.12065","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/2402.12065/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":"2402.12065","created_at":"2026-07-05T07:47:10.347436+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.12065v2","created_at":"2026-07-05T07:47:10.347436+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.12065","created_at":"2026-07-05T07:47:10.347436+00:00"},{"alias_kind":"pith_short_12","alias_value":"AHPKUNJ7TPD4","created_at":"2026-07-05T07:47:10.347436+00:00"},{"alias_kind":"pith_short_16","alias_value":"AHPKUNJ7TPD4BKWN","created_at":"2026-07-05T07:47:10.347436+00:00"},{"alias_kind":"pith_short_8","alias_value":"AHPKUNJ7","created_at":"2026-07-05T07:47:10.347436+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.08302","citing_title":"HACK++: Towards More Effective Head-Aware Key-Value Compression for Efficient Visual Autoregressive Modeling","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17757","citing_title":"OSCAR: Offline Spectral Covariance-Aware Rotation for 2-bit KV Cache Quantization","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2504.19874","citing_title":"TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate","ref_index":60,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12464","citing_title":"Search Your Block Floating Point Scales!","ref_index":168,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04738","citing_title":"OSAQ: Outlier Self-Absorption for Accurate Low-bit LLM Quantization","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2604.21072","citing_title":"Distributed Generative Inference of LLM at Internet Scales with Multi-Dimensional Communication Optimization","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04738","citing_title":"OSAQ: Outlier Self-Absorption for Accurate Low-bit LLM Quantization","ref_index":19,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AHPKUNJ7TPD4BKWNK26ZJHIG3T","json":"https://pith.science/pith/AHPKUNJ7TPD4BKWNK26ZJHIG3T.json","graph_json":"https://pith.science/api/pith-number/AHPKUNJ7TPD4BKWNK26ZJHIG3T/graph.json","events_json":"https://pith.science/api/pith-number/AHPKUNJ7TPD4BKWNK26ZJHIG3T/events.json","paper":"https://pith.science/paper/AHPKUNJ7"},"agent_actions":{"view_html":"https://pith.science/pith/AHPKUNJ7TPD4BKWNK26ZJHIG3T","download_json":"https://pith.science/pith/AHPKUNJ7TPD4BKWNK26ZJHIG3T.json","view_paper":"https://pith.science/paper/AHPKUNJ7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.12065&json=true","fetch_graph":"https://pith.science/api/pith-number/AHPKUNJ7TPD4BKWNK26ZJHIG3T/graph.json","fetch_events":"https://pith.science/api/pith-number/AHPKUNJ7TPD4BKWNK26ZJHIG3T/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AHPKUNJ7TPD4BKWNK26ZJHIG3T/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AHPKUNJ7TPD4BKWNK26ZJHIG3T/action/storage_attestation","attest_author":"https://pith.science/pith/AHPKUNJ7TPD4BKWNK26ZJHIG3T/action/author_attestation","sign_citation":"https://pith.science/pith/AHPKUNJ7TPD4BKWNK26ZJHIG3T/action/citation_signature","submit_replication":"https://pith.science/pith/AHPKUNJ7TPD4BKWNK26ZJHIG3T/action/replication_record"}},"created_at":"2026-07-05T07:47:10.347436+00:00","updated_at":"2026-07-05T07:47:10.347436+00:00"}