Shangding Gu

AccidentBench: Benchmarking Multimodal Understanding and Reasoning in Vehicle Accidents and Beyond

StyleBench: Evaluating thinking styles in Large Language Models

Data Uniformity Improves Training Efficiency and More, with a Convergence Framework Beyond the NTK Regime

RLBenchNet: Benchmarking Neural Architectures with PPO Across Reinforcement Learning Tasks

Don’t Trade Off Safety: Diffusion Regularization for Constrained Offline RL

Robust Gymnasium: A Unified Modular Benchmark for Robust Reinforcement Learning