7 papers
Gradient-Flow Optimization as Dynamic Random-Effects Inference: Testing and Early Stopping with Applications to Deep Learning
Minhao Yao, Ruoyu Wang, Xihong Lin +2
Gradient-flow optimization is usually viewed as an algorithmic procedure for minimizing empirical loss, with training duration selected by validation or heuristic early stopping ru…
Fitting Unknown Number of Hyperplanes with Manifold Optimization
Zhiqin Cheng, Yu Zhan, Mingjin Zhang +2
Fitting an unknown number of hyperplanes to data is a fundamental yet challenging problem in machine learning, characterized by its non-convexity, non-differentiability, and unknow…
SkiP: When to Skip and When to Refine for Efficient Robot Manipulation
Mingtong Dai, Guanqi Peng, Yongjie Bai +5
Previous imitation learning policies predict future actions at every control step, whether in smooth motion phases or precise, contact-rich operation phases. This uniform treatment…
Learning to See and Act: Task-Aware Virtual View Exploration for Robotic Manipulation
Yongjie Bai, Zhouxia Wang, Yang Liu +8
Recent vision-language-action (VLA) models for multi-task robot manipulation often rely on fixed camera setups and shared visual encoders, which limit their performance under occlu…
GraspView: Active Perception Scoring and Best-View Optimization for Robotic Grasping in Cluttered Environments
Shenglin Wang, Mingtong Dai, Jingxuan Su +4
Robotic grasping is a fundamental capability for autonomous manipulation, yet remains highly challenging in cluttered environments where occlusion, poor perception quality, and inc…
RoVer: Robot Reward Model as Test-Time Verifier for Vision-Language-Action Model
Mingtong Dai, Lingbo Liu, Yongjie Bai +6
Vision-Language-Action (VLA) models have become a prominent paradigm for embodied intelligence, yet further performance improvements typically rely on scaling up training data and…