PPL
Chengchun Shi
Papers (6)
BASIS: Batchwise Advantage Estimation from Single-Rollout Information Sharing for LLM ReasoningCounterfactually Safe Reinforcement LearningLearning Perturbations to Extrapolate Your LLMRobust Sequential Experimental Design for A/B TestingPerturbation is All You Need for Extrapolating Language ModelsSegmenting Human-LLM Co-authored Text via Change Point Detection