Xiaomeng Ye

Xiaomeng Ye is an undergraduate researcher in computer science at the Georgia Institute of Technology, where she works at the Robotics Perception and Learning (RIPL) Lab on sim-to-real reinforcement learning and vision-language-action model training using 3D Gaussian Splatting scene reconstructions. She is a lead author of the Splat2Act paper, which presents an automated real-to-sim-to-real pipeline for personalizing VLA manipulation policies to building-scale indoor environments using handheld phone scans. She has also contributed to the EmbodiedSplat project presented at ICCV 2025.