{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:QAYWPF7MPRHVLTWUXFAVQNCTWX","short_pith_number":"pith:QAYWPF7M","schema_version":"1.0","canonical_sha256":"80316797ec7c4f55ced4b941583453b5e8c67fca7087607436187f638aae5cff","source":{"kind":"arxiv","id":"2309.14717","version":2},"attestation_state":"computed","paper":{"title":"QA-LoRA: Quantization-Aware Low-Rank Adaptation of Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Heng Chang, Hengheng Zhang, Lingxi Xie, Qi Tian, Xiaopeng Zhang, Xiaotao Gu, Xin Chen, Yuhui Xu, Zhengsu Chen","submitted_at":"2023-09-26T07:22:23Z","abstract_excerpt":"Recently years have witnessed a rapid development of large language models (LLMs). Despite the strong ability in many language-understanding tasks, the heavy computational burden largely restricts the application of LLMs especially when one needs to deploy them onto edge devices. In this paper, we propose a quantization-aware low-rank adaptation (QA-LoRA) algorithm. The motivation lies in the imbalanced degrees of freedom of quantization and adaptation, and the solution is to use group-wise operators which increase the degree of freedom of quantization meanwhile decreasing that of adaptation. "},"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":"2309.14717","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-09-26T07:22:23Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"1c9f8fa96b8550f5db4426c5f8ecd276d6cd4179f06ea28b136d8c3a7cfcbef9","abstract_canon_sha256":"01b35669a1028d92e94f1c861776dc0f5c61772892946b07e59120d6fef33896"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:58:42.842034Z","signature_b64":"Jw9KD+Whf29FfVlPz2c5Hexl6yBxfOKM2Zwbq/DAnL6VWghT037Ggu1G+eoJvoGbhtk28KzoSRAv304S+W0WBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"80316797ec7c4f55ced4b941583453b5e8c67fca7087607436187f638aae5cff","last_reissued_at":"2026-07-05T06:58:42.841536Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:58:42.841536Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"QA-LoRA: Quantization-Aware Low-Rank Adaptation of Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Heng Chang, Hengheng Zhang, Lingxi Xie, Qi Tian, Xiaopeng Zhang, Xiaotao Gu, Xin Chen, Yuhui Xu, Zhengsu Chen","submitted_at":"2023-09-26T07:22:23Z","abstract_excerpt":"Recently years have witnessed a rapid development of large language models (LLMs). Despite the strong ability in many language-understanding tasks, the heavy computational burden largely restricts the application of LLMs especially when one needs to deploy them onto edge devices. In this paper, we propose a quantization-aware low-rank adaptation (QA-LoRA) algorithm. The motivation lies in the imbalanced degrees of freedom of quantization and adaptation, and the solution is to use group-wise operators which increase the degree of freedom of quantization meanwhile decreasing that of adaptation. "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.14717","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/2309.14717/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":"2309.14717","created_at":"2026-07-05T06:58:42.841601+00:00"},{"alias_kind":"arxiv_version","alias_value":"2309.14717v2","created_at":"2026-07-05T06:58:42.841601+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.14717","created_at":"2026-07-05T06:58:42.841601+00:00"},{"alias_kind":"pith_short_12","alias_value":"QAYWPF7MPRHV","created_at":"2026-07-05T06:58:42.841601+00:00"},{"alias_kind":"pith_short_16","alias_value":"QAYWPF7MPRHVLTWU","created_at":"2026-07-05T06:58:42.841601+00:00"},{"alias_kind":"pith_short_8","alias_value":"QAYWPF7M","created_at":"2026-07-05T06:58:42.841601+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":10,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.08194","citing_title":"Dive Into the Implicit Biases of Low-rank Vision-language Alignment","ref_index":48,"is_internal_anchor":true},{"citing_arxiv_id":"2606.01412","citing_title":"GPTQ-intrinsic LoRA: A Near-optimal Algorithm for Low-precision Quantization with Low-rank Adaptation","ref_index":71,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21854","citing_title":"CrossVLA: Cross-Paradigm Post-Training and Inference Optimization for Vision-Language-Action Models","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21854","citing_title":"CrossVLA: Cross-Paradigm Post-Training and Inference Optimization for Vision-Language-Action Models","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2602.04163","citing_title":"BPDQ: Bit-Plane Decomposition Quantization on a Variable Grid for Large Language Models","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20295","citing_title":"Quant.npu: Enabling Efficient Mobile NPU Inference for on-device LLMs via Fully Static Quantization","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2404.14294","citing_title":"A Survey on Efficient Inference for Large Language Models","ref_index":190,"is_internal_anchor":false},{"citing_arxiv_id":"2604.02501","citing_title":"ECG Foundation Models and Medical LLMs for Agentic Cardiovascular Intelligence at the Edge: A Review and Outlook","ref_index":100,"is_internal_anchor":false},{"citing_arxiv_id":"2403.14608","citing_title":"Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Survey","ref_index":127,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18655","citing_title":"Unlocking the Edge deployment and ondevice acceleration of multi-LoRA enabled one-for-all foundational LLM","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QAYWPF7MPRHVLTWUXFAVQNCTWX","json":"https://pith.science/pith/QAYWPF7MPRHVLTWUXFAVQNCTWX.json","graph_json":"https://pith.science/api/pith-number/QAYWPF7MPRHVLTWUXFAVQNCTWX/graph.json","events_json":"https://pith.science/api/pith-number/QAYWPF7MPRHVLTWUXFAVQNCTWX/events.json","paper":"https://pith.science/paper/QAYWPF7M"},"agent_actions":{"view_html":"https://pith.science/pith/QAYWPF7MPRHVLTWUXFAVQNCTWX","download_json":"https://pith.science/pith/QAYWPF7MPRHVLTWUXFAVQNCTWX.json","view_paper":"https://pith.science/paper/QAYWPF7M","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2309.14717&json=true","fetch_graph":"https://pith.science/api/pith-number/QAYWPF7MPRHVLTWUXFAVQNCTWX/graph.json","fetch_events":"https://pith.science/api/pith-number/QAYWPF7MPRHVLTWUXFAVQNCTWX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QAYWPF7MPRHVLTWUXFAVQNCTWX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QAYWPF7MPRHVLTWUXFAVQNCTWX/action/storage_attestation","attest_author":"https://pith.science/pith/QAYWPF7MPRHVLTWUXFAVQNCTWX/action/author_attestation","sign_citation":"https://pith.science/pith/QAYWPF7MPRHVLTWUXFAVQNCTWX/action/citation_signature","submit_replication":"https://pith.science/pith/QAYWPF7MPRHVLTWUXFAVQNCTWX/action/replication_record"}},"created_at":"2026-07-05T06:58:42.841601+00:00","updated_at":"2026-07-05T06:58:42.841601+00:00"}