Lavida-R1: Advancing Reasoning for Unified Multimodal Diffusion Language Models
Shufan Li, Yuchen Zhu, Jiuxiang Gu, Kangning Liu, Zhe Lin, Yongxin Chen, Molei Tao, Aditya Grover
Abstract
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.
BibTeX
@inproceedings{
li2026lavidar,
title={Lavida-R1: Advancing Reasoning for Unified Multimodal Diffusion Language Models},
author={Shufan Li and Yuchen Zhu and Kangning Liu and Zhe Lin and Yongxin Chen and Molei Tao and Aditya Grover and Jiuxiang Gu and Jason Kuen},
booktitle={Forty-third International Conference on Machine Learning},
year={2026},
url={https://openreview.net/forum?id=pVcEcaHHD2}
}