Xiaoyu Ma

Sampling rollouts via GRPO actively degrades tool-selection training as policies concentrate, whereas exact policy optimization over enumerable combinatorial tool spaces eliminates vanishing gradients and boosts accuracy by up to 14.2 points.

Incorporating visual feedback into prompt optimization leads to significant performance improvements in vision-language tasks, revealing previously unrecognized error patterns.

TARA redefines prompt optimization by routing specific failures to tailored repair operators, achieving superior semantic accuracy without the need for generator retraining.

You can slash LLM prompt evaluation costs by 35-60% without sacrificing accuracy by intelligently selecting which examples to use.