Xueqian Wang
4 papers · Latest:
Computer Vision
Power Reinforcement Post-Training of Text-to-Image Models with Super-Linear Advantage Shaping
SLAS improves text-to-image models by using a novel super-linear advantage shaping to mitigate reward hacking and enhance training efficiency and robustness.
2605.10937
RoboticsRPG: Robust Policy Gating for Smooth Multi-Skill Transitions in Humanoid Fighting
RPG enables humanoids to perform smooth, stable multi-skill fighting transitions by training a unified policy with randomization.
2604.21355
RoboticsLearn Weightlessness: Imitate Non-Self-Stabilizing Motions on Humanoid Robot
This paper introduces a Weightlessness Mechanism for humanoid robots to imitate non-self-stabilizing motions by adaptively relaxing joints for environmental interaction.
2604.21351
RoboticsOVAL: Open-Vocabulary Augmented Memory Model for Lifelong Object Goal Navigation
OVAL introduces a lifelong open-vocabulary memory model and a novel exploration strategy for efficient and robust object goal navigation in unseen environments.
2604.12872
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.