Shortcuts and Boundaries: Verifiable RL Misalignment and Contextual Safety Learning
Verifiable reward–context conflicts reveal how agentic RL learns shortcuts and how execution feedback supports useful, transferable contextual safety.
Xinjie Shen
Agentic RL
Shortcuts & Boundaries: RL Misalignment and Contextual Safety
How agentic RL can improve real work while learning contextual violations, and how verifiable feedback supports useful, transferable safety.
Verifiable Hidden Dynamics Play: Generating Agentic RL Environments from Solved Mechanisms
Generate diverse, stateful agentic RL environments with verifiable rewards by deriving both dynamics and evaluation from solved mechanisms.