OneLLM: One Framework to Align All Modalities with Language

Overall framework

Abstract

Multimodal large language models (MLLMs) have gained significant attention due to their strong multimodal understanding capability. However, existing works rely heavily on modality-specific encoders, which usually differ in architecture and are limited to common modalities. In this paper, we present OneLLM, an MLLM that aligns eight modalities to language using a unified framework. We achieve this through a unified multimodal encoder and a progressive multimodal alignment pipeline. In detail, we first train an image projection module to connect a vision encoder with LLM. Then, we build a universal projection module (UPM) by mixing multiple image projection modules and dynamic routing. Finally, we progressively align more modalities to LLM with the UPM. To fully leverage the potential of OneLLM in following instructions, we also curated a comprehensive multimodal instruction dataset, including 2M items from image, audio, video, point cloud, depth/normal map, IMU and fMRI brain activity. OneLLM is evaluated on 25 diverse benchmarks, encompassing tasks such as multimodal captioning, question answering and reasoning, where it delivers excellent performance.

Publication
CVPR 2024
Jiaqi Wang 王佳琦
Jiaqi Wang 王佳琦
Research Scientist & Team Leader
Shanghai AI Laboratory

Jiaqi Wang is currently a Research Scientist and Team Leader at Shanghai AI Laboratory, as well as an Adjunct Ph.D. Supervisor at Shanghai Jiao Tong University. His research focuses on visual perception, vision-language models, and the development of benchmarks and datasets in these areas.

Related