Datasets:

Modalities:
Image
Video
Size:
< 1K
ArXiv:
Libraries:
Datasets
License:
VIE-Bench / README.md
Simons212's picture
Update README.md
840f257 verified
---
license: apache-2.0
---
# VIE-Bench
We introduce VIE-Bench (Video Instruction-Based Editing Benchmark), proposed in [Instruct-X](https://github.com/MC-E/InstructX/), which comprises 140 high-quality instances across eight editing categories.
| Edit Task | Sub Edit Task | Number |
|:---------------------|:---------------------------:|------:|
| Total | | 140 |
| Local Edit | Object Swap | 25 |
| | Color Change | 10 |
| | Add | 30 |
| | Remove | 30 |
| Global Edit | Style Change | 10 |
| | Tone / Weather Change | 5 |
| Hybrid Edit | - | 10 |
| Reference Base Edit | Reference Base Swap | 10 |
| | Reference Base Add | 10 |
# InstructX
**[InstructX: Towards Unified Visual Editing with MLLM Guidance](https://arxiv.org/pdf/2510.08485)**
> [Chong Mou](https://scholar.google.com/citations?user=SYQoDk0AAAAJ&hl=zh-CN), [Qichao Sun](https://huggingface.co/Simons212), [Yanze Wu](https://scholar.google.com/citations?user=FdHiVvkAAAAJ&hl=zh-CN)<sup>&dagger;</sup><sup>&ddagger;</sup>, [Pengze Zhang](https://openreview.net/profile?id=%7EPengze_Zhang1), [Xinghui Li](https://crayon-shinchan.github.io/xinghui99.github.io/), [Fulong Ye](https://scholar.google.com/citations?user=-BbQ5VgAAAAJ&hl=zh-CN), [Songtao Zhao](https://openreview.net/profile?id=~Songtao_Zhao1)<sup>&ddagger;</sup>, [Qian He](https://scholar.google.com/citations?user=9rWWCgUAAAAJ)<br>
> <sup>&dagger;</sup> Corresponding author, <sup>&ddagger;</sup> Project lead;
> Intelligent Creation Lab, Bytedance
<p align="center">
<a href="https://mc-e.github.io/project/InstructX/">🌐 Project Page</a> |
<a href="https://arxiv.org/pdf/2510.08485">📜 Arxiv</a> |
<a href="https://huggingface.co/datasets/Simons212/VIE-Bench">🤗 Benchmark</a> |
</p>
### Updates
* **2025.10.10**: Release VIE-Bench.
* **2025.10.10**: Release InstructX tech report.