No-Chan Bae

GRIT: Gradient Masking for Visually Misgrounded Tokens in Text-Rich Reasoning

GRIT, a training framework that incorporates visual grounding into RLVR to mitigate the impact of Correct-but-Visually-Misgrounded Reasoning (CVMR), is proposed and introduced, and experiments demonstrate that GRIT significantly improves both reasoning accuracy and grounding fidelity over existing RLVR methods.