PPL
Weitong Zhang
Papers (5)
Moment Matching Q-LearningReturn-to-Go Is More Than a Number: Q-Guided Alignment for Return-Conditioned Supervised LearningWasserstein Equilibrium Decoding for Reliable Medical Visual Question AnsweringOGLS-SD: On-Policy Self-Distillation with Outcome-Guided Logit Steering for LLM ReasoningGeomHerd: A Forward-looking Herding Quantification via Ricci Flow Geometry on Agent Interactive Simulations