{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:EXCCOOUCH5LDI7WWPVP3IIWMNH","short_pith_number":"pith:EXCCOOUC","schema_version":"1.0","canonical_sha256":"25c4273a823f56347ed67d5fb422cc69f0499722464b8cf0fb85f0a6436a8e30","source":{"kind":"arxiv","id":"2501.13331","version":2},"attestation_state":"computed","paper":{"title":"Qrazor: Reliable and Effortless 4-bit LLM Quantization by Significant Data Razoring","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Dongyoung Lee, Ik Joon Chang, Seungkyu Choi","submitted_at":"2025-01-23T02:20:08Z","abstract_excerpt":"Large-scale language models (LLMs) excel in language processing tasks but face deployment challenges due to high memory and computational demands. While low-bit quantization, such as 4-bit techniques, offers a potential solution, these methods often suffer from significant accuracy loss or require considerable effort for implementation such as reordering, rotation, etc. To address these challenges, we propose QRazor, a simple yet effective quantization scheme that enables 4-bit quantization of weights, activations, and KV cache in transformer-based LLMs. QRazor operates in two stages: first, q"},"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":"2501.13331","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-23T02:20:08Z","cross_cats_sorted":[],"title_canon_sha256":"683de9122074dd5403cf9729ad4ce38cf9c12c68cc9759e4d5ec0557b4411aa5","abstract_canon_sha256":"5d6bedbfd2ca4d9fda2c26ff166571fb36cf9adc703d65c5a990889e6fec3e55"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:09:51.732548Z","signature_b64":"bRDncd4qQ8lRCnhsfHZI1vdeFwS1sJ3bxb7DZP4NJW1SYakxO2ciZ6isVgjjd+xzV5z9Zh0QCCHGQbzS88s7Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"25c4273a823f56347ed67d5fb422cc69f0499722464b8cf0fb85f0a6436a8e30","last_reissued_at":"2026-07-05T10:09:51.732075Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:09:51.732075Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Qrazor: Reliable and Effortless 4-bit LLM Quantization by Significant Data Razoring","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Dongyoung Lee, Ik Joon Chang, Seungkyu Choi","submitted_at":"2025-01-23T02:20:08Z","abstract_excerpt":"Large-scale language models (LLMs) excel in language processing tasks but face deployment challenges due to high memory and computational demands. While low-bit quantization, such as 4-bit techniques, offers a potential solution, these methods often suffer from significant accuracy loss or require considerable effort for implementation such as reordering, rotation, etc. To address these challenges, we propose QRazor, a simple yet effective quantization scheme that enables 4-bit quantization of weights, activations, and KV cache in transformer-based LLMs. QRazor operates in two stages: first, q"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.13331","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/2501.13331/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":"2501.13331","created_at":"2026-07-05T10:09:51.732129+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.13331v2","created_at":"2026-07-05T10:09:51.732129+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.13331","created_at":"2026-07-05T10:09:51.732129+00:00"},{"alias_kind":"pith_short_12","alias_value":"EXCCOOUCH5LD","created_at":"2026-07-05T10:09:51.732129+00:00"},{"alias_kind":"pith_short_16","alias_value":"EXCCOOUCH5LDI7WW","created_at":"2026-07-05T10:09:51.732129+00:00"},{"alias_kind":"pith_short_8","alias_value":"EXCCOOUC","created_at":"2026-07-05T10:09:51.732129+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/EXCCOOUCH5LDI7WWPVP3IIWMNH","json":"https://pith.science/pith/EXCCOOUCH5LDI7WWPVP3IIWMNH.json","graph_json":"https://pith.science/api/pith-number/EXCCOOUCH5LDI7WWPVP3IIWMNH/graph.json","events_json":"https://pith.science/api/pith-number/EXCCOOUCH5LDI7WWPVP3IIWMNH/events.json","paper":"https://pith.science/paper/EXCCOOUC"},"agent_actions":{"view_html":"https://pith.science/pith/EXCCOOUCH5LDI7WWPVP3IIWMNH","download_json":"https://pith.science/pith/EXCCOOUCH5LDI7WWPVP3IIWMNH.json","view_paper":"https://pith.science/paper/EXCCOOUC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.13331&json=true","fetch_graph":"https://pith.science/api/pith-number/EXCCOOUCH5LDI7WWPVP3IIWMNH/graph.json","fetch_events":"https://pith.science/api/pith-number/EXCCOOUCH5LDI7WWPVP3IIWMNH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EXCCOOUCH5LDI7WWPVP3IIWMNH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EXCCOOUCH5LDI7WWPVP3IIWMNH/action/storage_attestation","attest_author":"https://pith.science/pith/EXCCOOUCH5LDI7WWPVP3IIWMNH/action/author_attestation","sign_citation":"https://pith.science/pith/EXCCOOUCH5LDI7WWPVP3IIWMNH/action/citation_signature","submit_replication":"https://pith.science/pith/EXCCOOUCH5LDI7WWPVP3IIWMNH/action/replication_record"}},"created_at":"2026-07-05T10:09:51.732129+00:00","updated_at":"2026-07-05T10:09:51.732129+00:00"}