Paper Detail
Anda Cao, Zhuo Gou, Yi Wang, Kaixuan Chen, Yu Wang, Can Wang, Mingli Song, Jie Song
Merging multiple Low-Rank Adaptation (LoRA) experts into a single backbone is a promising approach for efficient multi-task deployment. While existing methods strive to alleviate interference via weight interpolation or subspace alignment, they rest upon the implicit assumption that all LoRA matrices contribute constructively to the merged model. In this paper, we uncover a critical bottleneck in current merging paradigms: the existence of $\textit{negative modules}$ -- specific LoRA layers that inherently degrade global performance upon merging. We propose $\textbf{E}$volutionary $\textbf{N}$egative $\textbf{M}$odule $\textbf{P}$runing ($\textbf{ENMP}$), a plug-and-play LoRA pruning method to locate and exclude these detrimental modules prior to merging. By leveraging an evolutionary search strategy, ENMP effectively navigates the discrete, non-differentiable landscape of module selection to identify optimal pruning configurations. Extensive evaluations demonstrate that ENMP consistently boosts the performance of existing merging algorithms, achieving a new state-of-the-art across both language and vision domains. Code is available at https://github.com/CaoAnda/ENMP-LoRAMerging.
No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.
No ranking explanation is available yet.
No tags.
@article{cao2026evolutionary,
title = {Evolutionary Negative Module Pruning for Better LoRA Merging},
author = {Anda Cao and Zhuo Gou and Yi Wang and Kaixuan Chen and Yu Wang and Can Wang and Mingli Song and Jie Song},
year = {2026},
abstract = {Merging multiple Low-Rank Adaptation (LoRA) experts into a single backbone is a promising approach for efficient multi-task deployment. While existing methods strive to alleviate interference via weight interpolation or subspace alignment, they rest upon the implicit assumption that all LoRA matrices contribute constructively to the merged model. In this paper, we uncover a critical bottleneck in current merging paradigms: the existence of \$\textbackslash{}textit\{negative modules\}\$ -- specific LoRA layers that},
url = {https://arxiv.org/abs/2604.17753},
keywords = {cs.AI, cs.CL, cs.CV},
eprint = {2604.17753},
archiveprefix = {arXiv},
}
{}