VisionLLM v2: An End-to-End Generalist Multimodal Large Language Model for Hundreds of Vision-Language Tasks
This work addresses the need for more versatile and efficient multimodal AI systems for researchers and practitioners, though it appears incremental by building on existing MLLM frameworks.
The paper tackles the problem of limited application scope in multimodal large language models by introducing VisionLLM v2, an end-to-end generalist model that unifies visual perception, understanding, and generation for hundreds of vision-language tasks, achieving performance comparable to task-specific models.
We present VisionLLM v2, an end-to-end generalist multimodal large model (MLLM) that unifies visual perception, understanding, and generation within a single framework. Unlike traditional MLLMs limited to text output, VisionLLM v2 significantly broadens its application scope. It excels not only in conventional visual question answering (VQA) but also in open-ended, cross-domain vision tasks such as object localization, pose estimation, and image generation and editing. To this end, we propose a new information transmission mechanism termed "super link", as a medium to connect MLLM with task-specific decoders. It not only allows flexible transmission of task information and gradient feedback between the MLLM and multiple downstream decoders but also effectively resolves training conflicts in multi-tasking scenarios. In addition, to support the diverse range of tasks, we carefully collected and combed training data from hundreds of public vision and vision-language tasks. In this way, our model can be joint-trained end-to-end on hundreds of vision language tasks and generalize to these tasks using a set of shared parameters through different user prompts, achieving performance comparable to task-specific models. We believe VisionLLM v2 will offer a new perspective on the generalization of MLLMs.