Update README.md
Browse files
README.md
CHANGED
@@ -7,7 +7,35 @@ sdk: static
|
|
7 |
pinned: false
|
8 |
---
|
9 |
|
10 |
-
# MJ-Bench Team
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
11 |
|
12 |
## π [**MJ-Video**: Fine-Grained Benchmarking and Rewarding Video Preferences in Video Generation](https://aiming-lab.github.io/MJ-VIDEO.github.io/)
|
13 |
|
@@ -39,7 +67,7 @@ We evaluate a wide range of multimodal judges, including:
|
|
39 |
- 4 closed-source VLMs (e.g., GPT-4, Claude 3)
|
40 |
|
41 |
<p align="center">
|
42 |
-
<img src="https://
|
43 |
</p>
|
44 |
|
45 |
π₯ **We are actively updating the [leaderboard](https://mj-bench.github.io/)!**
|
|
|
7 |
pinned: false
|
8 |
---
|
9 |
|
10 |
+
# MJ-Bench Team
|
11 |
+
|
12 |
+
[MJ-Bench-Team](https://mj-bench.github.io/) is co-founded by Stanford University, UNC-Chapel Hill, and the University of Chicago. We aim to align modern foundation models with multimodal judges to enhance reliability, safety, and performance.
|
13 |
+
|
14 |
+
<p align="center">
|
15 |
+
<img
|
16 |
+
src="https://raw.githubusercontent.com/MJ-Bench/MJ-Bench.github.io/main/static/images/Stanford.jpg"
|
17 |
+
alt="Stanford University"
|
18 |
+
width="350"
|
19 |
+
style="display:inline-block; margin:0 40px; vertical-align:middle;"/>
|
20 |
+
<img
|
21 |
+
src="https://raw.githubusercontent.com/MJ-Bench/MJ-Bench.github.io/main/static/images/UNC-logo.png"
|
22 |
+
alt="UNC Chapel Hill"
|
23 |
+
width="190"
|
24 |
+
style="display:inline-block; margin:0 -10px; vertical-align:middle;"/>
|
25 |
+
<img
|
26 |
+
src="https://raw.githubusercontent.com/MJ-Bench/MJ-Bench.github.io/main/static/images/UChicago-logo.jpg"
|
27 |
+
alt="University of Chicago"
|
28 |
+
width="160"
|
29 |
+
style="display:inline-block; margin:0 140px; vertical-align:middle;"/>
|
30 |
+
</p>
|
31 |
+
|
32 |
+
---
|
33 |
+
|
34 |
+
## Recent News
|
35 |
+
- π **MJ-PreferGen** is **accepted by ICLR25**! Check out the paper: [*MJ-PreferGen: An Automatic Framework for Preference Data Synthesis*](https://openreview.net/forum?id=WpZyPk79Fu)
|
36 |
+
- π₯ We have released [**MJ-Video**](https://aiming-lab.github.io/MJ-VIDEO.github.io/). All datasets and model checkpoints are available [here](https://huggingface.co/MJ-Bench)!
|
37 |
+
|
38 |
+
---
|
39 |
|
40 |
## π [**MJ-Video**: Fine-Grained Benchmarking and Rewarding Video Preferences in Video Generation](https://aiming-lab.github.io/MJ-VIDEO.github.io/)
|
41 |
|
|
|
67 |
- 4 closed-source VLMs (e.g., GPT-4, Claude 3)
|
68 |
|
69 |
<p align="center">
|
70 |
+
<img src="https://raw.githubusercontent.com/MJ-Bench/MJ-Bench/main/assets/overview_new.png" alt="MJ-Bench Dataset Overview" width="80%"/>
|
71 |
</p>
|
72 |
|
73 |
π₯ **We are actively updating the [leaderboard](https://mj-bench.github.io/)!**
|