Eyes in the Update? Learning Where Rewards Land in Multimodal Reasoningwith Lixuan He, Shulin Tian, Shikang Zheng, Feng Chen +11 versionW_wvpszucuv1 · current