V-Rubrics

New V-Rubrics method enhances vision-language model grounding

Researchers have developed V-Rubrics, a novel reinforcement learning approach to improve the visual faithfulness and reasoning consistency of vision-language models. This method decomposes reference responses into atomi…