Models for physical AI tasks, such as robotics and autonomous driving, need not reason solely in language. They can use visual plans--points, bounding boxes, and trajectories--to represent spatial ...

Reliable robotic manipulation requires both semantic understanding and precise control in three-dimensional space. However, vision-language-action (VLA) foundations largely inherit visual ...

Streaming video understanding requires models to repeatedly decide whether to respond or continue observing. Recently, Group Relative Policy Optimization (GRPO) has been applied to this setting, with ...

Harness evolution from a common base exhibits Locality, Complementarity, and Nonstationarity: variants repair different failures, and their utility changes as the policy learns. Rather than seeking a ...

Learning a universal action representation from video requires modeling dynamics that are consistent across different executions of the same action yet invariant to changes in embodiment and ...

暂无中文译文，以下为英文原文。点「开始对话」，AI 可以用中文给你讲解这篇论文。 On-policy distillation (OPD) trains a student on ...

Benchmark compression reduces the often prohibitive time and cost of model evaluation by evaluating models on a small benchmark subset that closely matches full-benchmark results. To quantify ...

暂无中文译文，以下为英文原文。点「开始对话」，AI 可以用中文给你讲解这篇论文。 Vision-Language-Action (VLA) models have made ...

暂无中文译文，以下为英文原文。点「开始对话」，AI 可以用中文给你讲解这篇论文。 This paper introduces Quaternion Approximate ...

Unified conditional image generation requires a single model to preserve different aspects of a visual input while producing a target image that follows textual guidance. Pretrained video models ...

暂无中文译文，以下为英文原文。点「开始对话」，AI 可以用中文给你讲解这篇论文。 Outcome RL raises pass@1 but not pass@k at large k ...

Vision-language models are shifting from offline passive understanding toward online proactive interaction. However, prevailing methods largely rely on explicit requests while miss the implicit user ...