D2E: Scaling Vision-Action Pretraining on Desktop Data for Transfer to Embodied AI
Paper • 2510.05684 • Published • 147
Xingchao Liu
XCLiu
AI & ML interests
None yet
Recent Activity
authored a paper about 8 hours ago
Flow Straight and Fast: Learning to Generate and Transfer Data with
Rectified Flow upvoted a paper 9 months ago
D2E: Scaling Vision-Action Pretraining on Desktop Data for Transfer to
Embodied AI authored a paper over 1 year ago
TripoSG: High-Fidelity 3D Shape Synthesis using Large-Scale Rectified
Flow ModelsOrganizations
upvoted a paper 9 months ago