Papers › DoRA: Weight-Decomposed Low-Rank Adaptation

DoRA: Weight-Decomposed Low-Rank Adaptation

14 Feb 2024arXiv:2402.09353archive 2025-07-28

Shih-Yang Liu, Chien-Yi Wang, Hongxu Yin, Pavlo Molchanov, Yu-Chiang Frank Wang, Kwang-Ting Cheng, Min-Hung Chen

Among the widely used parameter-efficient fine-tuning (PEFT) methods, LoRA and its variants have gained considerable popularity because of avoiding additional inference costs. However, there still often exists an accuracy gap between these methods and full fine-tuning (FT). In this work, we first introduce a novel weight decomposition analysis to investigate the inherent differences between FT and LoRA. Aiming to resemble the learning capacity of FT from the findings, we propose Weight-Decomposed Low-Rank Adaptation (DoRA). DoRA decomposes the pre-trained weight into two components, magnitude and direction, for fine-tuning, specifically employing LoRA for directional updates to efficiently minimize the number of trainable parameters. By employing \ours, we enhance both the learning capacity and training stability of LoRA while avoiding any additional inference overhead. \ours~consistently outperforms LoRA on fine-tuning LLaMA, LLaVA, and VL-BART on various downstream tasks, such as commonsense reasoning, visual instruction tuning, and image/video-text understanding. Code is available at https://github.com/NVlabs/DoRA.

PaperPDFCodeCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2402.09353")

Code

Syntology Ran 8 of 15 code samples harvested from 2 repositories linked to this paper; 7 have no recorded run. Of those that ran: 6 ran · our draft was wrong; 2 ran with no contract checked.

By repository: official repository: 14 samples from 1 repository, 7 ran; community (archive-listed): 1 sample from 1 repository, 1 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

NVlabs/DoRA officialmentioned in papermentioned on GitHubpytorchNOASSERTION report
nbasyl/DoRA officialmentioned in papermentioned on GitHubNOASSERTION report
catid/dora mentioned on GitHubpytorch report
seanzhang-zhichen/llama3-chinese mentioned on GitHubpytorchApache-2.0 report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

15 samples harvested; 8 ran; 0 honoured the contract we drafted; 7 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

6ran · our draft was wrong
2ran
7unverified

Licence: 14 of the 15 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 2 repositories linked to this paper, official or community; each sample names its own and says which. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: 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. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

collate_fn NVlabs/DoRA/image_video_text_understanding/feature_extraction/refcocog_mattnet.py official repository ran no licence file found · pointer only · 68e41c28a6dfe81d · report
evaluate NVlabs/DoRA/commonsense_reasoning/multi_dataset_eval.py official repository ran fingerprinted no licence file found · pointer only · 2dc17f5e6fb39b3f · report
generate_prompt NVlabs/DoRA/commonsense_reasoning/finetune.py official repository ran · our draft was wrong no licence file found · pointer only · 56e674752db00649 · report
generate_prompt NVlabs/DoRA/commonsense_reasoning/generate.py official repository ran · our draft was wrong fingerprinted no licence file found · pointer only · 0dbc7fc6c518d0fb · report
generate_prompt NVlabs/DoRA/commonsense_reasoning/commonsense_evaluate.py official repository ran · our draft was wrong fingerprinted no licence file found · pointer only · 1c6b8be890b6118a · report
load_data NVlabs/DoRA/commonsense_reasoning/commonsense_evaluate.py official repository ran · our draft was wrong no licence file found · pointer only · 56a9466ff35b35bb · report
translate_state_dict_key NVlabs/DoRA/commonsense_reasoning/export_state_dict_checkpoint.py official repository ran · our draft was wrong no licence file found · pointer only · a4f5d0414a1cf9c6 · report
get_peft_state_maybe_zero_3 NVlabs/DoRA/visual_instruction_tuning/llava/train/train_dora.py official repository unverified licence not identified · pointer only · 0d95286fcf19dd47 · report
get_peft_state_non_dora_maybe_zero_3 NVlabs/DoRA/visual_instruction_tuning/llava/train/train_dora.py official repository unverified licence not identified · pointer only · 293ec4b71aa63222 · report
main NVlabs/DoRA/commonsense_reasoning/commonsense_evaluate.py official repository unverified no licence file found · pointer only · 415f46d8a341e28d · report
main NVlabs/DoRA/commonsense_reasoning/evaluate.py official repository unverified no licence file found · pointer only · 3398969fc7af6ff7 · report
maybe_zero_3 NVlabs/DoRA/visual_instruction_tuning/llava/train/train_dora.py official repository unverified no licence file found · pointer only · 616ffbdc154ed2d8 · report
permute NVlabs/DoRA/commonsense_reasoning/export_state_dict_checkpoint.py official repository unverified no licence file found · pointer only · a2f49e13f6e13d6b · report
unpermute NVlabs/DoRA/commonsense_reasoning/export_state_dict_checkpoint.py official repository unverified no licence file found · pointer only · 6e275f232bbfb9b6 · report
generate_data catid/dora/dora.py community (archive-listed) ran · our draft was wrong fingerprinted MIT (permissive) · f040823d10747c45 · report

Tasks

parameter-efficient fine-tuning

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
parameter-efficient fine-tuning BoolQ LLaMA2-7b Accuracy (% ) 81.93 #3 of 4 Archive leaderboard report
parameter-efficient fine-tuning HellaSwag LLaMA2-7b Accuracy (% ) 76.27 #3 of 3 Archive leaderboard report
parameter-efficient fine-tuning WinoGrande LLaMA2-7b Accuracy (% ) 70.09 #2 of 3 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections