… However, OPD training becomes unstable when the teacher and student distributions differ substantially, as teacher supervision on student-generated tokens may yield unreliable policy gradients and even cause optimization failure. …
… AI-generated summary Distillation-based acceleration has become foundational for making autoregressive streaming video diffusion models practical, with distribution matching distillation DMD as the de facto choice. …
Papers arxiv:2605.01402 Injecting Distributional Awareness into MLLMs via Reinforcement Learning for Deep Imbalanced Regression Published on May 11 Submitted by DUYao on May 12 Authors: Yao Du , , Abstract A distribution-aware reinforcement learning framework improves multimodal large language mode… …
… Mainstream methods such as PPO and GRPO approximate this control with a ratio-clipping mechanism, but the importance ratio can be a poor proxy for distributional shift in long-tailed vocabularies. …
… Spatial Adaptive Multi Guidance in Diffusion Models 2026 Diffusion-APO: Trajectory-Aware Direct Preference Alignment for Video Diffusion Transformers 2026 Diffusion Image Generation with Explicit Modeling of Data Manifold Geometry 2026 Continuous-Time Distribution Matching for Few-Step Diffusion Di… …
… Existing counting models are often tailored to scenarios such as crowds, vehicles, cells, crops, or remote-sensing objects, and thus struggle to generalize across categories, visual domains, object scales, and density distributions. …
… Published on May 14 Submitted by JingyeChen22 on May 15 HKUST Authors: , Haolin Yang , , , Jingye Chen , Abstract Synthetic layered image data improves graphic design decomposition by enabling scalable training and better layer distribution control compared to traditional methods. …
… Distilling such generators from high-fidelity bidirectional teachers yields competitive few-step models, yet a persistent gap between the history distributions encountered during training and those arising at inference constrains generation quality over long horizons. …
…Generated by Qwen/Qwen2.5-Coder-32B-Instruct Visual reasoning requires integrating evidence distributed across regions, attributes, and relations, making single-chain reasoning prone to early perceptual commitment and hallucination . We propose…
… We validate this across diverse simulation and real-world setups, with gains reaching +22.5% under out-of-distribution scenarios. …