LaViDa-R1: Advancing Reasoning for Unified Multimodal Diffusion Language Models
This work addresses the need for unified reasoning models in multimodal AI, though it appears incremental by building on existing diffusion language models.
The authors tackled the problem of building a general-purpose reasoning diffusion language model for multimodal tasks by proposing LaViDa-R1, which achieved strong performance on diverse tasks like visual math reasoning and image editing through a unified post-training framework.
Diffusion language models (dLLMs) recently emerged as a promising alternative to auto-regressive LLMs. The latest works further extended it to multimodal understanding and generation tasks. In this work, we propose LaViDa-R1, a multimodal, general-purpose reasoning dLLM. Unlike existing works that build reasoning dLLMs through task-specific reinforcement learning, LaViDa-R1 incorporates diverse multimodal understanding and generation tasks in a unified manner. In particular, LaViDa-R1 is built with a novel unified post-training framework that seamlessly integrates supervised finetuning (SFT) and multi-task reinforcement learning (RL). It employs several novel training techniques, including answer-forcing, tree search, and complementary likelihood estimation, to enhance effectiveness and scalability. Extensive experiments demonstrate LaViDa-R1's strong performance on a wide range of multimodal tasks, including visual math reasoning, reason-intensive grounding, and image editing.