AI RESEARCH

Diffusion Large Language Models for Visual Speech Recognition

arXiv CS.AI

ArXi:2605.28456v1 Announce Type: new Existing Visual Speech Recognition (VSR) systems commonly rely on left-to-right autoregressive decoding, which can force premature decisions on visually ambiguous tokens before sufficient context is available. We propose DLLM-VSR, to the best of our knowledge, the first Diffusion Large Language Model (DLLM)-based VSR framework, formulating transcription as iterative masked denoising with flexible-order decoding.