{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:5LCHSYTWB5K4VJ7LSVZFUKTLCL","short_pith_number":"pith:5LCHSYTW","schema_version":"1.0","canonical_sha256":"eac47962760f55caa7eb95725a2a6b12f415e63fa297e374e52f74835c5831ac","source":{"kind":"arxiv","id":"2407.11534","version":2},"attestation_state":"computed","paper":{"title":"LRQ: Optimizing Post-Training Quantization for Large Language Models by Learning Low-Rank Weight-Scaling Matrices","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Dongsoo Lee, Eunho Yang, Jeonghoon Kim, June Yong Yang, Jung Hyun Lee, Kang Min Yoo, Se Jung Kwon","submitted_at":"2024-07-16T09:32:07Z","abstract_excerpt":"With the commercialization of large language models (LLMs), weight-activation quantization has emerged to compress and accelerate LLMs, achieving high throughput while reducing inference costs. However, existing post-training quantization (PTQ) techniques for quantizing weights and activations of LLMs still suffer from non-negligible accuracy drops, especially on massive multitask language understanding. To address this issue, we propose Low-Rank Quantization (LRQ) - a simple yet effective post-training weight quantization method for LLMs that reconstructs the outputs of an intermediate Transf"},"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":"2407.11534","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-07-16T09:32:07Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c2a0010a2395315a2a422e159d3ad13685a3dadaae57a71445709a5922fd37ae","abstract_canon_sha256":"d038c491a8eb3329ab99db03f7585f6c3831c2bb24a17bfefa54acdb7bf9f2ed"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:11:22.222756Z","signature_b64":"QExbiJAsWEcS4hWaGG72J92XhXx+TlEn5dK5ZjYn7ixUlpUfC0PXCuRd3Gxzu60bSXpvV2p9ann0zvjiCeGnBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"eac47962760f55caa7eb95725a2a6b12f415e63fa297e374e52f74835c5831ac","last_reissued_at":"2026-07-05T10:11:22.222151Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:11:22.222151Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LRQ: Optimizing Post-Training Quantization for Large Language Models by Learning Low-Rank Weight-Scaling Matrices","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Dongsoo Lee, Eunho Yang, Jeonghoon Kim, June Yong Yang, Jung Hyun Lee, Kang Min Yoo, Se Jung Kwon","submitted_at":"2024-07-16T09:32:07Z","abstract_excerpt":"With the commercialization of large language models (LLMs), weight-activation quantization has emerged to compress and accelerate LLMs, achieving high throughput while reducing inference costs. However, existing post-training quantization (PTQ) techniques for quantizing weights and activations of LLMs still suffer from non-negligible accuracy drops, especially on massive multitask language understanding. To address this issue, we propose Low-Rank Quantization (LRQ) - a simple yet effective post-training weight quantization method for LLMs that reconstructs the outputs of an intermediate Transf"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.11534","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/2407.11534/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":"2407.11534","created_at":"2026-07-05T10:11:22.222218+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.11534v2","created_at":"2026-07-05T10:11:22.222218+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.11534","created_at":"2026-07-05T10:11:22.222218+00:00"},{"alias_kind":"pith_short_12","alias_value":"5LCHSYTWB5K4","created_at":"2026-07-05T10:11:22.222218+00:00"},{"alias_kind":"pith_short_16","alias_value":"5LCHSYTWB5K4VJ7L","created_at":"2026-07-05T10:11:22.222218+00:00"},{"alias_kind":"pith_short_8","alias_value":"5LCHSYTW","created_at":"2026-07-05T10:11:22.222218+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.06534","citing_title":"ROSE: Rollout On Serving GPUs via Cooperative Elasticity for Agentic RL","ref_index":33,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5LCHSYTWB5K4VJ7LSVZFUKTLCL","json":"https://pith.science/pith/5LCHSYTWB5K4VJ7LSVZFUKTLCL.json","graph_json":"https://pith.science/api/pith-number/5LCHSYTWB5K4VJ7LSVZFUKTLCL/graph.json","events_json":"https://pith.science/api/pith-number/5LCHSYTWB5K4VJ7LSVZFUKTLCL/events.json","paper":"https://pith.science/paper/5LCHSYTW"},"agent_actions":{"view_html":"https://pith.science/pith/5LCHSYTWB5K4VJ7LSVZFUKTLCL","download_json":"https://pith.science/pith/5LCHSYTWB5K4VJ7LSVZFUKTLCL.json","view_paper":"https://pith.science/paper/5LCHSYTW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.11534&json=true","fetch_graph":"https://pith.science/api/pith-number/5LCHSYTWB5K4VJ7LSVZFUKTLCL/graph.json","fetch_events":"https://pith.science/api/pith-number/5LCHSYTWB5K4VJ7LSVZFUKTLCL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5LCHSYTWB5K4VJ7LSVZFUKTLCL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5LCHSYTWB5K4VJ7LSVZFUKTLCL/action/storage_attestation","attest_author":"https://pith.science/pith/5LCHSYTWB5K4VJ7LSVZFUKTLCL/action/author_attestation","sign_citation":"https://pith.science/pith/5LCHSYTWB5K4VJ7LSVZFUKTLCL/action/citation_signature","submit_replication":"https://pith.science/pith/5LCHSYTWB5K4VJ7LSVZFUKTLCL/action/replication_record"}},"created_at":"2026-07-05T10:11:22.222218+00:00","updated_at":"2026-07-05T10:11:22.222218+00:00"}