paper-with-me

홈 › Papers

Can LoRA Fusion Support Cross-Domain Tasks in Cloud-Edge Collaboration?

2026-04-09 · Yatong Wang, Fali Wang, Naibin Gu, Zheng Lin, Zhengxiao Liu, Dingyu Yao, Zhiwei Zhang, Jianxin Shi, Weiping Wang arxiv

Cloud-hosted large language models (LLMs) commonly rely on LoRA for domain adaptation, yet domain data are distributed across multiple edge devices and cannot be uploaded due to privacy constraints. This raises a fundamental question: how can knowledge from multiple private edges be integrated into a cloud LLM for cross-domain problem solving? A natural solution is to train LoRA adapters locally and fuse them in the cloud; however, existing pipelines rely on unrealistic assumptions that edge devices can host cloud-scale LLMs and are evaluated mainly on single-domain tasks. To address these limitations, we propose a prune-train-recover framework that enables local LoRA training on pruned models and privacy-preserving cloud integration. We further introduce MMLU-CD, a cross-domain benchmark that composes multiple domain samples into a single instance, enabling explicit evaluation of cross-domain problem solving. This allows us to ask a concrete question: Can existing LoRA fusion methods support cross-domain tasks in cloud-edge collaboration? Our empirical answer is negative. Existing LoRA fusion methods perform poorly on MMLU-CD, often underperforming the base LLM, revealing their inability to support cross-domain problem solving. We attribute this failure to parameter conflicts among LoRA adapters and propose a simple conflict-resolution module, LoRA-CR, which mitigates conflicting updates and improves LoRA fusion performance by up to 3.8%. These results identify conflict mitigation as a critical yet largely overlooked factor in cloud-edge LoRA fusion, warranting further investigation in future research.

📄 PDF Abstract BibTeX arXiv:2605.23913

Code (0)

등록된 구현이 없습니다.

Tasks

Domain Adaptation

Similar Papers 제목 키워드 기반

qa-FLoRA: Data-free query-adaptive Fusion of LoRAs for LLMs

2025-12-12 · Shreya Shukla, Aditya Sriram, Milinda Kuppur Narayanaswamy, Hiteshi Jain arxiv

The deployment of large language models for specialized tasks often requires domain-specific parameter-efficient finetuning through Low-Rank Adaptation (LoRA) modules. However, effectively fusing these adapters to handle…

Domain Adaptation

ICM-Fusion: In-Context Meta-Optimized LoRA Fusion for Multi-Task Adaptation

2025-08-06 · Yihua Shao, Xiaofeng Lin, Xinwei Long, Siyu Chen 외 arxiv

Enabling multi-task adaptation in pre-trained Low-Rank Adaptation (LoRA) models is crucial for enhancing their generalization capabilities. Most existing pre-trained LoRA fusion methods decompose weight matrices, sharing…

Incremental Learning

LoRA-Flow: Dynamic LoRA Fusion for Large Language Models in Generative Tasks

2024-02-18 · Hanqing Wang, Bowen Ping, Shuo Wang, Xu Han 외

LoRA employs lightweight modules to customize large language models (LLMs) for each downstream task or domain, where different learned additional modules represent diverse skills. Combining existing LoRAs to address new …

Math

Adaptive Cyclic Diffusion for Inference Scaling

2025-05-20 · Gyubin Lee, Truong Nhat Nguyen Bao, Jaesik Yoon, Dongwoo Lee 외

Diffusion models have demonstrated strong generative capabilities across domains ranging from image synthesis to complex reasoning tasks. However, most inference-time scaling methods rely on fixed denoising schedules, li…

Computational EfficiencyDenoisingImage Generation

Mix-of-Show: Decentralized Low-Rank Adaptation for Multi-Concept Customization of Diffusion Models

2023-05-29 · NeurIPS 2023 11 · YuChao Gu, Xintao Wang, Jay Zhangjie Wu, Yujun Shi 외

Public large-scale text-to-image diffusion models, such as Stable Diffusion, have gained significant attention from the community. These models can be easily customized for new concepts using low-rank adaptations (LoRAs)…

Attribute