AI RESEARCH
Stream-R1: Reliability-Perplexity Aware Reward Distillation for Streaming Video Generation
arXiv CS.CV
•
ArXi:2605.03849v1 Announce Type: new Distillation-based acceleration has become foundational for making autoregressive streaming video diffusion models practical, with distribution matching distillation (DMD) as the de facto choice. Existing methods, however, train the student to match the teacher's output indiscriminately, treating every rollout, frame, and pixel as equally reliable supervision.