Mengdi Wang

7 papers Princeton University

• CURE: Co-Evolving Coders and Unit Testers via Reinforcement Learning
• DISC: Dynamic Decomposition Improves LLM Inference Scaling
• Does Thinking More Always Help? Mirage of Test-Time Scaling in Reasoning Models
• MMaDA: Multimodal Large Diffusion Language Models
• ReasonFlux-PRM: Trajectory-Aware PRMs for Long Chain-of-Thought Reasoning in LLMs
• Securing the Language of Life: Inheritable Watermarks from DNA Language Models to Proteins
• Training-Free Guidance Beyond Differentiability: Scalable Path Steering with Tree Search in Diffusion and Flow Models