AI.info

Xinbing Wang

Explore Xinbing Wang on AI.info.

• Flow of Spans: Generalizing Language Models to Dynamic Span-Vocabulary via GFlowNets
• Extreme Value Policy Optimization for Safe Reinforcement Learning
• Controlling Underestimation Bias in Constrained Reinforcement Learning for Safe Exploration
• AWM: Accurate Weight-Matrix Fingerprint for Large Language Models