model distillation

5 papers

• Antidistillation Sampling
• Generative Data Augmentation via Diffusion Distillation, Adversarial Alignment, and Importance Reweighting
• Preference Distillation via Value based Reinforcement Learning
• ReasonFlux-PRM: Trajectory-Aware PRMs for Long Chain-of-Thought Reasoning in LLMs
• VarFlow: Proper Scoring-Rule Diffusion Distillation via Energy Matching