AI RESEARCH
Visual Latents Know More Than They Say: Unsilencing Latent Reasoning in MLLMs
arXiv CS.LG
•
ArXi:2605.02735v1 Announce Type: new Continuous latent-space reasoning offers a compact alternative to textual chain-of-thought for multimodal models, enabling high-dimensional visual evidence to be integrated without explicit reasoning tokens. However, we identify a previously overlooked optimization pathology in existing latent visual reasoning methods: although visual latents become semantically enriched during