Siyuan Huang
Senior researcher and corresponding author overseeing the LARA research direction.
“TeleDexter trains a single RL controller that handles in-hand translation, rotation, finger gaiting, and long-horizon tool use”
Source→“TeleDexter achieves 75.2% average SR and 87.1% average TP across all seven tasks, while all baselines near-uniformly fail”
Source→“Siyuan Huang, BIGAI, Corresponding author. Leads research at the intersection of dexterous manipulation, simulation, and learning”
Source→“State Key Lab of General Artificial Intelligence at BIGAI. Co-authored ManipTrans... a sustained research program on dexterous grasping and manipulation”
Source→“LARA: Latent Action Representation Alignment for Vision-Language-Action Models”
Source→“LARA's core bet is that you can squeeze more performance out of existing unlabeled video (human or robot) by jointly training a Latent Action Model (LAM) and a diffusion-based VLA policy, rather than treating them as separate sequential steps.”
Source→AI-extracted from podcast / newsletter / paper summaries. May contain errors.