PPL
Masashi Sugiyama
Papers (6)
Accelerated Dynamic Importance Weighting with Versatile Divergence-Minimizing EstimatorsMahjax: A GPU-Accelerated Mahjong Simulator for Reinforcement Learning in JAXEmbracing Biased Transition Matrices for Complementary-Label Learning with Many ClassesShaping Schema via Language Representation as the Next Frontier for LLM Intelligence ExpandingDecomposing the Basic Abilities of Large Language Models: Mitigating Cross-Task Interference in Multi-Task Instruct-TuningData-dependent Exploration for Online Reinforcement Learning from Human Feedback