Mixture of Cluster-conditional LoRA Experts for Vision-language Instruction Tuning
Instruction tuning of Large Vision-language Models (LVLMs) has revolutionized the development of versatile models with zero-shot generalization across a wide range of downstream vision-language tasks. However, the diversity of training tasks of different sources and formats would lead to inevitable task conflicts, where different tasks conflict for the same set of model parameters, resulting in sub-optimal instructionfollowing abilities. To address that, we propose the Mixture of Clusterconditional LoRA Experts (MoCLE), a novel Mixture of Experts (MoE) architecture designed to activate the task-customized model parameters based on the instruction clusters. A separate universal expert is further incorporated to improve generalization capabilities of MoCLE for novel instructions. Extensive experiments on 11 zero-shot tasks demonstrate the effectiveness of MoCLE.
PDF Abstract
Visual Question Answering
Flickr30k
GQA
OK-VQA
TextVQA
Hateful Memes
ScienceQA
A-OKVQA
VSR
IconQA
iVQA
