2 Matching Annotations
  1. Last 7 days
  2. arxiv.org arxiv.org
    Med-R1: Reinforcement Learning for Generalizable Medical Reasoning in Vision-Language Models
    2
    2. NILAY1999 20 Dec 2025
      However, Think-After still performsslightly below No-Thinking, indicating that additional mecha-nisms—such as residual contextual coupling or reasoning-tokennoise—may also contribute to the performance gap

      imp

    Visit annotations in context

    Annotators

    URL

    arxiv.org/pdf/2503.13939