Zhang-Wei Hong

A Reward-Free Viewpoint on Multi-Objective Reinforcement Learning

BOAD: Discovering Hierarchical Software Engineering Agents via Bandit Optimization

Composition-Grounded Data Synthesis for Visual Reasoning

EvoScale: Evolutionary Test-Time Scaling for Software Engineering

Tailored Primitive Initialization is the Secret Key to Reinforcement Learning

Large Pretraining Datasets Don't Guarantee Robustness after Fine-Tuning

GPTOpt: Towards Efficient LLM-based Black-Box Optimization

RL Tango: Reinforcing Generator and Verifier Together for Language Reasoning

ORSO: Accelerating Reward Design via Online Reward Selection and Policy Optimization

Satori: Reinforcement Learning with Chain-of-Action-Thought Enhances LLM Reasoning via Autoregressive Search

ReGen: Generative Robot Simulation via Inverse Design

ImageNet-RIB Benchmark: Large Pre-Training Datasets Don't Guarantee Robustness after Fine-Tuning