Predicting temporal progress from visual trajectories is important for intelligent robots that can learn, adapt, and improve. However, learning such progress estimator, or temporal value function, across different tasks and domains requires both a large amount of diverse data and methods which can scale and generalize. To address these challenges, we present Generative Value Learning (\GVL), a universal value function estimator that leverages the world knowledge embedded in vision-language models (VLMs) to predict task progress. Naively asking a VLM to predict values for a video sequence performs poorly due to the strong temporal correlation between successive frames. Instead, GVL poses value estimation as a temporal ordering problem over shuffled video frames; this seemingly more challenging task encourages VLMs to more fully exploit their underlying semantic and temporal grounding capabilities to differentiate frames based on their perceived task progress, consequently producing significantly better value predictions. Without any robot or task specific training, GVL can in-context zero-shot and few-shot predict effective values for more than 300 distinct real-world tasks across diverse robot platforms, including challenging bimanual manipulation tasks. Furthermore, we demonstrate that GVL permits flexible multi-modal in-context learning via examples from heterogeneous tasks and embodiments, such as human videos. The generality of GVL enables various downstream applications pertinent to visuomotor policy learning, including dataset filtering, success detection, and advantage-weighted regression -- all without any model training or finetuning.
PrivacyMind: Large Language Models Can Be Contextual Privacy Protection Learners
University of California, Los Angeles（加州大学洛杉矶分校）；Georgia Institute of Technology（佐治亚理工学院）；Tsinghua University（清华大学）；University of California, Santa Barbara（加州大学圣巴巴拉分校）；NEC Laboratories America（美国NEC实验室）
From Learner Behavior to Reusable Skills for Effective and Efficient Learner Simulation
National University of Defense Technology（国防科技大学）；Nanyang Technological University（南洋理工大学）；The Hong Kong Polytechnic University（香港理工大学）；Southeast University（东南大学）
Predictive Inverse Dynamics Models are Scalable Learners for Robotic Manipulation
Shanghai AI Laboratory（上海人工智能实验室）；School of Software & Microelectronics, Peking University（北京大学软件与微电子学院）；Chinese University of Hong Kong（香港中文大学）；National Engineering Research Center for Software Engineering, Peking University（北京大学国家软件工程研究中心）；CFCS, School of CS, Peking University（北京大学计算机科学学院CFCS）
Lifelong Learner: Discovering Versatile Neural Solvers for Vehicle Routing Problems
department of Computer Science at National Yang Ming Chiao Tung University（国立阳明交通大学计算机科学系）；Institute for Infocomm Research, Agency for Science, Technology and Research (A*STAR)（资讯通信研究院，科技研究局（A*STAR））；College of Computing and Data Science at Nanyang Technological University（南洋理工大学计算与数据科学学院）；School of Computer Science, Sichuan Normal University（四川师范大学计算机科学学院）；School of Computer Science and Engineering, Nanyang Technological University（南洋理工大学计算机科学与工程学院）；Center for Frontier AI Research, Agency for Science, Technology and Research (A*STAR)（前沿人工智能研究中心，科技研究局（A*STAR））
