Yao Zhou
Yao Zhou is a researcher affiliated with the Chinese Academy of Sciences who works on vision-language-action reinforcement learning and multimodal AI systems. He is known for identifying and addressing trajectory-level credit aliasing in outcome-driven VLA reinforcement learning through his Temporal GRPO framework, which constructs detectable task stages and applies stage-specific advantages to improve policy training. His broader research spans vision-language pre-training, geo-localization with reasoning, and efficient embedding model optimization.
“We characterize and formulate trajectory-level credit aliasing in outcome-driven VLA reinforcement learning, where rollouts with different stage progress can receive the same final-outcome advantage, causing successful preceding actions and later failed actions to be updated with the same advantage.”
Source→AI-extracted from podcast / newsletter / paper summaries. May contain errors.