PPL
Shuze Daniel Liu
Papers (4)
Offline Two-Player Zero-Sum Markov Games with KL RegularizationPredicting Plasticity in Deep Continual Learning: A Theoretical PerspectiveBeyond Linear Attention: Softmax Transformers Implement In-Context Reinforcement LearningAstroAlertBench: Evaluating the Accuracy, Reasoning, and Honesty of Multimodal LLMs in Astronomical Classification