This repository contains the datasets used in the paper SimpleMemVLA: A Simple but Effective Native-Video Memory for Vision-Language-Action Models. The datasets are provided in LeRobot v3 format and include per-frame sub-task annotations, camera observations, actions, and proprioceptive state for long-horizon robotic manipulation benchmarks.
The datasets cover the following benchmarks: RMBench, RoboMME, MIKASA-Robo, RoboMemArena, and LIBERO. For download instructions, training, and evaluation details, see the GitHub repository.
This repository contains the datasets used in the paper SimpleMemVLA: A Simple but Effective Native-Video Memory for Vision-Language-Action Models. The datasets are provided in LeRobot v3 format and include per-frame sub-task annotations, camera observations, actions, and proprioceptive state for long-horizon robotic manipulation benchmarks.
The datasets cover the following benchmarks: RMBench, RoboMME, MIKASA-Robo, RoboMemArena, and LIBERO. For download instructions, training, and evaluation details, see the GitHub repository.
