@inproceedings{huang2024mixture,
author = {Huang, Shaohan and Wei, Furu},
title = {Mixture of LoRA Experts},
booktitle = {ICLR 2024},
year = {2024},
month = {April},
abstract = {Low-Rank Adaptation (LoRA) has emerged as a pivotal technique for fine-tuning large pre-trained models, renowned for its efficacy across a wide array of tasks. The modular architecture of LoRA has catalyzed further research into the synergistic composition of multiple trained LoRAs, aiming to amplify performance across various tasks. However, the effective composition of these trained LoRAs presents a formidable challenge: (1) Linear arithmetic composition can lead to the diminution of the generative capabilities inherent in the original pre-trained models or the distinctive attributes of the individually trained LoRAs, potentially resulting in suboptimal outcomes. (2) Reference tuning-based composition exhibits limitations in adaptability and incurs significant computational costs due to the requirements to retrain a large model. In response to these challenges, we propose Mixture of LoRA Experts (MOLE). MOLE treats each layer of trained LoRAs as a distinct expert and implements hierarchical weight control by integrating a learnable gating function within each layer to learn optimal  composition weights tailored specifically to the objectives of a given domain. MOLE not only demonstrates enhanced performance in LoRA composition but also preserves the essential flexibility necessary for effective composition of trained LoRAs with minimal computational overhead. Extensive experiments conducted in both Natural Language Processing (NLP) and Vision & Language (V&L) domains validate the effects of MOLE.},
url = {http://approjects.co.za/?big=en-us/research/publication/mixture-of-lora-experts/},
}