{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:ADZMJGORXQSIAAQX7SYAECDTNX","short_pith_number":"pith:ADZMJGOR","schema_version":"1.0","canonical_sha256":"00f2c499d1bc24800217fcb00208736dcadfa9c2f63fdd5aa26e1512b262a5e9","source":{"kind":"arxiv","id":"2306.02272","version":4},"attestation_state":"computed","paper":{"title":"OWQ: Outlier-Aware Weight Quantization for Efficient Fine-Tuning and Inference of Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Changhun Lee, Eunhyeok Park, Hyungjun Kim, Jungyu Jin, Taesu Kim","submitted_at":"2023-06-04T06:33:13Z","abstract_excerpt":"Large language models (LLMs) with hundreds of billions of parameters require powerful server-grade GPUs for inference, limiting their practical deployment. To address this challenge, we introduce the outlier-aware weight quantization (OWQ) method, which aims to minimize LLM's footprint through low-precision representation. OWQ prioritizes a small subset of structured weights sensitive to quantization, storing them in high-precision, while applying highly tuned quantization to the remaining dense weights. This sensitivity-aware mixed-precision scheme reduces the quantization error notably, and "},"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":"2306.02272","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-04T06:33:13Z","cross_cats_sorted":[],"title_canon_sha256":"cfc24096dc763820714f2c687026fa2bd7314786b72d50965e7b1f3670ef582a","abstract_canon_sha256":"956b2ab40cc88169a9836176994cefed36493eb0b0b43aff08083c84aa8ffe44"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:36:51.533344Z","signature_b64":"ze2vXMEOw05NO3x00aq+JzJDUCwgnugKVeJnR9ButoxHO2Bq914FNdczc/BBQ8vT5BkORNer1612J2Fz1rU5Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"00f2c499d1bc24800217fcb00208736dcadfa9c2f63fdd5aa26e1512b262a5e9","last_reissued_at":"2026-07-05T07:36:51.532751Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:36:51.532751Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"OWQ: Outlier-Aware Weight Quantization for Efficient Fine-Tuning and Inference of Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Changhun Lee, Eunhyeok Park, Hyungjun Kim, Jungyu Jin, Taesu Kim","submitted_at":"2023-06-04T06:33:13Z","abstract_excerpt":"Large language models (LLMs) with hundreds of billions of parameters require powerful server-grade GPUs for inference, limiting their practical deployment. To address this challenge, we introduce the outlier-aware weight quantization (OWQ) method, which aims to minimize LLM's footprint through low-precision representation. OWQ prioritizes a small subset of structured weights sensitive to quantization, storing them in high-precision, while applying highly tuned quantization to the remaining dense weights. This sensitivity-aware mixed-precision scheme reduces the quantization error notably, and "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.02272","kind":"arxiv","version":4},"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/2306.02272/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":"2306.02272","created_at":"2026-07-05T07:36:51.532830+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.02272v4","created_at":"2026-07-05T07:36:51.532830+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.02272","created_at":"2026-07-05T07:36:51.532830+00:00"},{"alias_kind":"pith_short_12","alias_value":"ADZMJGORXQSI","created_at":"2026-07-05T07:36:51.532830+00:00"},{"alias_kind":"pith_short_16","alias_value":"ADZMJGORXQSIAAQX","created_at":"2026-07-05T07:36:51.532830+00:00"},{"alias_kind":"pith_short_8","alias_value":"ADZMJGOR","created_at":"2026-07-05T07:36:51.532830+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.18800","citing_title":"Theory-optimal Quantization Based on Flatness","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2307.06435","citing_title":"A Comprehensive Overview of Large Language Models","ref_index":259,"is_internal_anchor":false},{"citing_arxiv_id":"2404.14294","citing_title":"A Survey on Efficient Inference for Large Language Models","ref_index":195,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12358","citing_title":"Why and When Visual Token Pruning Fails? A Study on Relevant Visual Information Shift in MLLMs Decoding","ref_index":24,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ADZMJGORXQSIAAQX7SYAECDTNX","json":"https://pith.science/pith/ADZMJGORXQSIAAQX7SYAECDTNX.json","graph_json":"https://pith.science/api/pith-number/ADZMJGORXQSIAAQX7SYAECDTNX/graph.json","events_json":"https://pith.science/api/pith-number/ADZMJGORXQSIAAQX7SYAECDTNX/events.json","paper":"https://pith.science/paper/ADZMJGOR"},"agent_actions":{"view_html":"https://pith.science/pith/ADZMJGORXQSIAAQX7SYAECDTNX","download_json":"https://pith.science/pith/ADZMJGORXQSIAAQX7SYAECDTNX.json","view_paper":"https://pith.science/paper/ADZMJGOR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.02272&json=true","fetch_graph":"https://pith.science/api/pith-number/ADZMJGORXQSIAAQX7SYAECDTNX/graph.json","fetch_events":"https://pith.science/api/pith-number/ADZMJGORXQSIAAQX7SYAECDTNX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ADZMJGORXQSIAAQX7SYAECDTNX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ADZMJGORXQSIAAQX7SYAECDTNX/action/storage_attestation","attest_author":"https://pith.science/pith/ADZMJGORXQSIAAQX7SYAECDTNX/action/author_attestation","sign_citation":"https://pith.science/pith/ADZMJGORXQSIAAQX7SYAECDTNX/action/citation_signature","submit_replication":"https://pith.science/pith/ADZMJGORXQSIAAQX7SYAECDTNX/action/replication_record"}},"created_at":"2026-07-05T07:36:51.532830+00:00","updated_at":"2026-07-05T07:36:51.532830+00:00"}