Embodied AI TopConf · ICLR2026

D2E: Scaling Vision-Action Pretraining on Desktop Data for Transfer to Embodied AI

ICLR2026 / Benchmark and Dataset

Benchmark/数据集感知数据
新标签打开 AlphaXiv
D2E: Scaling Vision-Action Pretraining on Desktop Data for Transfer to Embodied AI figure
AlphaXiv 中文论文页面（可滚动查看）