PPL
Tian Lan
Papers (8)
FedQHD: Closed-Form Function-Space Federated Reinforcement LearningOPPO: Bayesian Value Recursion for Token-Level Credit Assignment in LLM ReasoningImproving Multi-turn Dialogue Consistency with Self-Recall ThinkingCase-Based Calibration of Adaptive Reasoning and Execution for LLM Tool UseMatrix-Space Reinforcement Learning for Reusing Local Transition GeometryMetric-Gradient Projection for Stable Multi-Agent Policy LearningInteractive Critique-Revision Training for Reliable Structured LLM GenerationOperator-Guided Invariance Learning for Continuous Reinforcement Learning