7 papers
KungfuAthleteBot: learning high-dynamic humanoid motion from video with unified robust recovery
Zhongxiang Lei, Lulu Cao, Xuyang Wang +3
Video is an abundant, inexpensive source of human motion data that is rich in extreme athletic behaviors. Making it usable for humanoid robots, however, is not a matter of simply r…
ForgetMimic: Motion Unlearning for Reinforcement Learning Humanoid Control
Xukun Luan, Zhongxiang Lei, Chen Gong +3
Humanoid control, leveraging human demonstrations, has achieved diverse, agile, and natural locomotion behaviors through reinforcement learning (RL). While this paradigm has yielde…
VLALeaks: Membership Inference Attacks against Vision-Language-Action Models
Xukun Luan, Jinyan Liu, Xuesong Li +4
Vision-Language-Action (VLA) models enable end-to-end robot control and have garnered widespread attention. However, the memorization of training data inherent to VLA, coupled with…
Creative Convergence or Imitation? Genre-Specific Homogeneity in LLM-Generated Chinese Literature
Yuanchi Ma, Kaize Shi, Hui He +5
Large Language Models (LLMs) have demonstrated remarkable capabilities in narrative generation. However, they often produce structurally homogenized stories, frequently following r…
KungfuAthleteBot: Learning High-Dynamic Motion from Videos with Unified Robust Recovery
Zhongxiang Lei, Lulu Cao, Xuyang Wang +3
Learning highly dynamic motions from video provides a scalable alternative to expensive data collection for humanoid robots, yet it is challenged by height floating, missing kinema…
Why Federated Optimization Fails to Achieve Perfect Fitting? A Theoretical Perspective on Client-Side Optima
Zhongxiang Lei, Qi Yang, Ping Qiu +3
Federated optimization is a constrained form of distributed optimization that enables training a global model without directly sharing client data. Although existing algorithms can…
Masked AutoEncoder for Graph Clustering without Pre-defined Cluster Number k
Yuanchi Ma, Hui He, Zhongxiang Lei +1
Graph clustering algorithms with autoencoder structures have recently gained popularity due to their efficient performance and low training cost. However, for existing graph autoen…