| license: mit | |
| task_categories: | |
| - robotics | |
| tags: | |
| - lerobot | |
| - mobile-manipulation | |
| - vla | |
| # EBench: Elemental Mobile Manipulation Benchmark Dataset | |
| [Project Page](https://internrobotics.github.io/EBench-home/) | [Paper](https://huggingface.co/papers/2606.18239) | [GitHub](https://github.com/InternRobotics/EBench) | [Documentation](https://internrobotics.github.io/EBench-doc/) | |
| EBench is an indoor VLA (Vision-Language-Action) manipulation benchmark built on NVIDIA Isaac Sim. Instead of compressing a model's behavior into a single overall success rate, it produces a **multi-axis capability profile** that exposes what a model is good at — and where it overfits. | |
| This repository contains the training trajectories (stored in the [LeRobot](https://github.com/huggingface/lerobot) format) used in the EBench evaluation suite. | |
| ## Key Features | |
| - **Three manipulation regimes in one benchmark** — covers *long-horizon*, *dexterous & precise*, and *mobile* manipulation. | |
| - **5-axis atomic diagnostic** — every task is labelled by *Scene · Atomic Skill · Horizon · Precision · Mobility*. | |
| - **4-axis generalization tests** — controlled perturbations along *Object · Background · Instruction · Mixed*. | |
| - **Strict train / test isolation** — validation and unseen splits are open for tuning, while a held-out test split drives the leaderboard. | |
| ## Citation | |
| ```bibtex | |
| @misc{ebench2026, | |
| title = {EBench: Elemental Mobile Manipulation Benchmark}, | |
| author = {Shanghai AI Laboratory}, | |
| year = {2026}, | |
| note = {Preprint coming soon}, | |
| url = {https://internrobotics.github.io/EBench-doc/} | |
| } | |
| ``` |