{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/lora-pro-are-low-rank-adapters-properly","title":"LoRA-Pro: Are Low-Rank Adapters Properly Optimized?","arxiv_id":"2407.18242","date":"2024-07-25","proceeding":null,"authors":["Zhengbo Wang","Jian Liang","Ran He","Zilei Wang","Tieniu Tan"],"abstract":"Low-rank adaptation, also known as LoRA, has emerged as a prominent method for parameter-efficient fine-tuning of foundation models. Despite its computational efficiency, LoRA still yields inferior performance compared to full fine-tuning. In this paper, we first uncover a fundamental connection between the optimization processes of LoRA and full fine-tuning: using LoRA for optimization is mathematically equivalent to full fine-tuning using a low-rank gradient for parameter updates. And this low-rank gradient can be expressed in terms of the gradients of the two low-rank matrices in LoRA. Leveraging this insight, we introduce LoRA-Pro, a method that enhances LoRA's performance by strategically adjusting the gradients of these low-rank matrices. This adjustment allows the low-rank gradient to more accurately approximate the full fine-tuning gradient, thereby narrowing the performance gap between LoRA and full fine-tuning. Furthermore, we theoretically derive the optimal solutions for adjusting the gradients of the low-rank matrices, applying them during fine-tuning in LoRA-Pro. We conduct extensive experiments across natural language understanding, dialogue generation, mathematical reasoning, code generation, and image classification tasks, demonstrating that LoRA-Pro substantially improves LoRA's performance, effectively narrowing the gap with full fine-tuning. Code is publicly available at https://github.com/mrflogs/LoRA-Pro.","url_abs":"https://arxiv.org/abs/2407.18242v3","url_pdf":"https://arxiv.org/pdf/2407.18242v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"lora-pro-are-low-rank-adapters-properly","repo_url":"https://github.com/mrflogs/LoRA-Pro","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"code-generation","task_name":"Code Generation"},{"task_slug":"computational-efficiency","task_name":"Computational Efficiency"},{"task_slug":"dialogue-generation","task_name":"Dialogue Generation"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"mathematical-reasoning","task_name":"Mathematical Reasoning"},{"task_slug":"natural-language-understanding","task_name":"Natural Language Understanding"},{"task_slug":"image-classification","task_name":"image-classification"},{"task_slug":"parameter-efficient-fine-tuning","task_name":"parameter-efficient fine-tuning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2407.18242","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2407.18242"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"deterministic:regex_extraction","url":"https://github.com/mrflogs/LoRA-Pro","reach":null}],"summary":{"ran_violates":1,"unverified":1},"by_repo_kind":{"official":{"samples":2,"ran":1,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"ab7439dcc2482e78","entry":"isclose","repo":"mrflogs/LoRA-Pro","repo_kind":"official","path":"DeepSpeed-0.15.1/deepspeed/runtime/zero/stage_1_and_2.py","file_url":"https://github.com/mrflogs/LoRA-Pro/blob/HEAD/DeepSpeed-0.15.1/deepspeed/runtime/zero/stage_1_and_2.py","link_basis":"first_harvest_node","language":"python","status":"ran_violates","verification_level":2,"contract_check":"VIOLATES","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"ab7439dcc2482e78"}},{"code_sha256_prefix":"29e02b1391f2ac78","entry":"lcm","repo":"mrflogs/LoRA-Pro","repo_kind":"official","path":"DeepSpeed-0.15.1/deepspeed/runtime/zero/stage_1_and_2.py","file_url":"https://github.com/mrflogs/LoRA-Pro/blob/HEAD/DeepSpeed-0.15.1/deepspeed/runtime/zero/stage_1_and_2.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"29e02b1391f2ac78"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}