AI RESEARCH

LaRe: Latent Refocusing for Multimodal Reasoning

arXiv CS.CL

ArXi:2511.02360v4 Announce Type: replace-cross Chain of Thought (CoT) reasoning enhances logical performance by decomposing complex tasks, yet its multimodal extension faces a trade-off. The prevailing Thinking with Images paradigm achieves visual refocusing by explicitly cropping image regions, yet incurs rapidly growing computational overhead. The emerging line of latent-space reasoning reduces token consumption, but lacks the capacity for dynamic refocusing.