Shizhe Diao

4 papers NVIDIA

• GPAS: Accelerating Convergence of LLM Pretraining via Gradient-Preserving Activation Scaling
• Nemotron-CLIMB: Clustering-based Iterative Data Mixture Bootstrapping for Language Model Pre-training
• Nemotron-Flash: Towards Latency-Optimal Hybrid Small Language Models
• ProRL: Prolonged Reinforcement Learning Expands Reasoning Boundaries in Large Language Models