File size: 1,741 Bytes
2079909
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
---
license: mit
task_categories:
- robotics
tags:
- lerobot
- vla
- mobile-manipulation
---

# EBench: Elemental Mobile Manipulation Benchmark

This repository contains the training trajectories (stored in LeRobot format) for **EBench**, 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, EBench produces a multi-axis capability profile that exposes what a model is good at and where it overfits.

- **Paper:** [EBench: Elemental Diagnosis of Generalist Mobile Manipulation Policies](https://huggingface.co/papers/2606.18239)
- **Project Page:** [EBench Homepage](https://internrobotics.github.io/EBench-home/)
- **Documentation:** [EBench Docs](https://internrobotics.github.io/EBench-doc/)
- **GitHub Repository:** [InternRobotics/EBench](https://github.com/InternRobotics/EBench)

## 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*.
- **LeRobot Format** — trajectories are structured using Hugging Face's `lerobot` data format, making them ready for loading and training.

## Citation

If you find EBench useful in your research, please cite:

```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/}
}
```