Saved in:
Bibliographic Details
Main Authors: Liu, Haokun, Je, Gyung Hyun, Ciccone, Marco, Xu, Zhenlin, YSS, Prasanth, Raffel, Colin
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.12323
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911444447002624
author Liu, Haokun
Je, Gyung Hyun
Ciccone, Marco
Xu, Zhenlin
YSS, Prasanth
Raffel, Colin
author_facet Liu, Haokun
Je, Gyung Hyun
Ciccone, Marco
Xu, Zhenlin
YSS, Prasanth
Raffel, Colin
contents The widespread availability of fine-tuned LoRA modules for open pre-trained models has led to an interest in methods that can adaptively merge LoRAs to improve performance. These methods typically include some way of selecting LoRAs from a pool and tune merging coefficients based on a task-specific dataset. While adaptive merging methods have demonstrated improvements in some settings, no past work has attempted to recycle LoRAs found "in the wild" on model repositories like the Hugging Face Hub. To address this gap, we consider recycling from a pool of nearly 1,000 user-contributed LoRAs trained from the Llama 3.1 8B-Instruct language model. Our empirical study includes a range of adaptive and non-adaptive merging methods in addition to a new method designed via a wide search over the methodological design space. We demonstrate that adaptive merging methods can improve performance over the base model but provide limited benefit over training a new LoRA on the same data used to set merging coefficients. We additionally find not only that the specific choice of LoRAs to merge has little importance, but that using LoRAs with randomly initialized parameter values yields similar performance. This raises the possibility that adaptive merging from recycled LoRAs primarily works via some kind of regularization effect, rather than by enabling positive cross-task transfer. To better understand why past work has proven successful, we confirm that positive transfer is indeed possible when there are highly relevant LoRAs in the pool. We release the model checkpoints and code online.
format Preprint
id arxiv_https___arxiv_org_abs_2602_12323
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle The Appeal and Reality of Recycling LoRAs with Adaptive Merging
Liu, Haokun
Je, Gyung Hyun
Ciccone, Marco
Xu, Zhenlin
YSS, Prasanth
Raffel, Colin
Machine Learning
Software Engineering
I.2.7
The widespread availability of fine-tuned LoRA modules for open pre-trained models has led to an interest in methods that can adaptively merge LoRAs to improve performance. These methods typically include some way of selecting LoRAs from a pool and tune merging coefficients based on a task-specific dataset. While adaptive merging methods have demonstrated improvements in some settings, no past work has attempted to recycle LoRAs found "in the wild" on model repositories like the Hugging Face Hub. To address this gap, we consider recycling from a pool of nearly 1,000 user-contributed LoRAs trained from the Llama 3.1 8B-Instruct language model. Our empirical study includes a range of adaptive and non-adaptive merging methods in addition to a new method designed via a wide search over the methodological design space. We demonstrate that adaptive merging methods can improve performance over the base model but provide limited benefit over training a new LoRA on the same data used to set merging coefficients. We additionally find not only that the specific choice of LoRAs to merge has little importance, but that using LoRAs with randomly initialized parameter values yields similar performance. This raises the possibility that adaptive merging from recycled LoRAs primarily works via some kind of regularization effect, rather than by enabling positive cross-task transfer. To better understand why past work has proven successful, we confirm that positive transfer is indeed possible when there are highly relevant LoRAs in the pool. We release the model checkpoints and code online.
title The Appeal and Reality of Recycling LoRAs with Adaptive Merging
topic Machine Learning
Software Engineering
I.2.7
url https://arxiv.org/abs/2602.12323