PPL
Jie M. Zhang
Papers (4)
Reasoning-Trace Collapse: Evaluating the Loss of Explicit Reasoning During Fine-TuningBeyond Execution: Static-Analysis Rewards and Hint-Conditioned Diffusion RL for Code GenerationStepCodeReasoner: Aligning Code Reasoning with Stepwise Execution Traces via Reinforcement LearningThe Attacker in the Mirror: Breaking Self-Consistency in Safety via Anchored Bipolicy Self-Play