…directing optimization capacity to the most informative learning signal. Furthermore, it estimates an adaptive per-problem target length online based on the model's own correct rollouts, applying a symmetric efficiency reward…
…Yanzuo Lu , , Abstract RAVEN enables real-time video generation through causal autoregressive extrapolation with improved training alignment, while CM-GRPO enhances performance via reinforcement learning applied to consistency model sampling. AI-generated…
…those whose answers never appear in 128 pre-RL samples (only ~18% of training data) drive ~83% of the gain, since rare correct rollouts still emerge during training and get reinforced. Together…
…Starting from a pretrained VLA policy, LWD closes the loop between deployment, shared physical experience, policy improvement , and redeployment by using autonomous rollouts and human interventions collected across a robot fleet. To…
…Trained primarily for short-horizon continuation, they treat long sequences as extended single shots, inevitably suffering from motion stagnation and semantic drift during long rollouts. To bridge this gap, we introduce CausalCine…
…Rather than relying on a static recipe, EvoTrainer enables LLM policies and their training harnesses to evolve jointly over time. This is more than conventional AI development, it is AI evolution in…
…AI-generated summary On-policy distillation (OPD) is a powerful paradigm for model alignment , yet its reliance on teacher logits restricts its application to white-box scenarios. We contend that structured semantic…
…existing reinforcement learning approaches in both training efficiency and final performance. AI-generated summary Reinforcement learning has emerged as a powerful tool for improving diffusion-based text-to-image models, but existing…
…reduced latency and improved quality compared to existing chunk-wise approaches. AI-generated summary Real-time interactive video generation requires low-latency, streaming, and controllable rollout. Existing autoregressive (AR) diffusion distillation methods…
…when all sampled rollouts for a query fail, the relative advantage collapses to zero. Consequently, the model loses effective training signals for these questions, wasting the training data and computational budget. While…