AI RESEARCH

Masked Diffusion Vision-Language Models for Temporal Action Localization

arXiv CS.CV

ArXi:2605.29858v1 Announce Type: new Temporal action localization (TAL) requires recognizing the target event and localizing its start and end times precisely in untrimmed videos. Recent vision-language formulations improve semantic reasoning and language-conditioned outputs, but their autoregressive decoders still generate tokens from left to right, preventing later semantic evidence from revising earlier timestamp predictions.