Hugging Face's logo

Spaces:
UI-MOPD
/
README
Configuration error

App Files Community
EliSpctre commited on
Commit
e774587
·
verified ·
1 Parent(s): fcca305

Fix model sizes (32B/8B), add thinking tag, add all datasets and eval results

Browse files
Files changed (1) hide show
  1. index.html +73 -62
index.html CHANGED
@@ -1,62 +1,73 @@
1
- <!DOCTYPE html>
2
- <html>
3
- <head>
4
- <meta charset="utf-8">
5
- <style>
6
- body { font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', Roboto, sans-serif; line-height: 1.6; color: #333; max-width: 800px; margin: 0 auto; padding: 20px; }
7
- .title { text-align: center; margin-bottom: 30px; }
8
- .title h1 { margin-bottom: 0; font-size: 1.8em; }
9
- .title h1 span { color: #FF6900; }
10
- .title p { margin-top: 4px; color: #555; font-size: 1.05em; }
11
- h3 { border-bottom: 1px solid #eee; padding-bottom: 6px; margin-top: 30px; }
12
- table { width: 100%; border-collapse: collapse; margin: 15px 0; }
13
- th, td { text-align: left; padding: 8px 12px; border: 1px solid #e0e0e0; }
14
- th { background: #f7f7f7; font-weight: 600; }
15
- a { color: #2563eb; text-decoration: none; }
16
- a:hover { text-decoration: underline; }
17
- ul { padding-left: 20px; }
18
- li { margin: 6px 0; }
19
- </style>
20
- </head>
21
- <body>
22
-
23
- <div class="title">
24
- <h1><span>UI-MOPD</span>: Multi-platform On-Policy Distillation for Continual GUI Agent Learning</h1>
25
- </div>
26
-
27
- <p>
28
- We build cross-platform GUI agents that can operate both desktop and mobile interfaces through a unified training framework.
29
- </p>
30
-
31
- <h3>Research</h3>
32
- <p>UI-MOPD introduces a two-stage training pipeline:</p>
33
- <ul>
34
- <li><b>Stage 1</b>: Supervised Fine-Tuning (SFT) on platform-specific teacher models</li>
35
- <li><b>Stage 2</b>: Reinforcement Learning distillation (DAPO) with multi-teacher on-policy guidance</li>
36
- </ul>
37
- <p>Our student model (8B) learns from multiple 32B teacher models to achieve strong cross-platform GUI interaction capabilities.</p>
38
-
39
- <h3>Models</h3>
40
- <table>
41
- <tr><th>Model</th><th>Size</th><th>Description</th></tr>
42
- <tr><td><a href="https://huggingface.co/UI-MOPD/Qwen3-VL-32B-Thinking-Desktop-Teacher">Qwen3-VL-32B-Thinking-Desktop-Teacher</a></td><td>33B</td><td>Desktop platform teacher</td></tr>
43
- <tr><td><a href="https://huggingface.co/UI-MOPD/Qwen3-VL-32B-Thinking-Mobile-Teacher">Qwen3-VL-32B-Thinking-Mobile-Teacher</a></td><td>33B</td><td>Mobile platform teacher</td></tr>
44
- <tr><td><a href="https://huggingface.co/UI-MOPD/Qwen3-VL-8B-Thinking-Desktop-SFT">Qwen3-VL-8B-Thinking-Desktop-SFT</a></td><td>9B</td><td>Desktop SFT checkpoint</td></tr>
45
- <tr><td><a href="https://huggingface.co/UI-MOPD/Qwen3-VL-8B-Thinking-Mobile-SFT">Qwen3-VL-8B-Thinking-Mobile-SFT</a></td><td>9B</td><td>Mobile SFT checkpoint</td></tr>
46
- <tr><td><a href="https://huggingface.co/UI-MOPD/Qwen3-VL-8B-Thinking-UI-MOPD-Student">Qwen3-VL-8B-Thinking-UI-MOPD-Student</a></td><td>9B</td><td>Final cross-platform student</td></tr>
47
- </table>
48
-
49
- <h3>Datasets</h3>
50
- <table>
51
- <tr><th>Dataset</th><th>Description</th></tr>
52
- <tr><td><a href="https://huggingface.co/datasets/UI-MOPD/Uni-GUI-OpenCUA">Uni-GUI-OpenCUA</a></td><td>Post-processed desktop trajectories from OpenCUA (~832 episodes, ~14K steps)</td></tr>
53
- <tr><td><a href="https://huggingface.co/datasets/UI-MOPD/Uni-GUI-Desktop-1">Uni-GUI-Desktop-1</a></td><td>Large-scale desktop GUI trajectories (~2.7K episodes, ~36K steps)</td></tr>
54
- </table>
55
-
56
- <h3>Links</h3>
57
- <ul>
58
- <li><a href="https://elispctre.github.io/UI-MOPD">Project Page</a></li>
59
- </ul>
60
-
61
- </body>
62
- </html>
 
 
 
 
 
 
 
 
 
 
 
 
1
+ <!DOCTYPE html>
2
+ <html>
3
+ <head>
4
+ <meta charset="utf-8">
5
+ <style>
6
+ body { font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', Roboto, sans-serif; line-height: 1.6; color: #333; max-width: 800px; margin: 0 auto; padding: 20px; }
7
+ .title { text-align: center; margin-bottom: 30px; }
8
+ .title h1 { margin-bottom: 0; font-size: 1.8em; }
9
+ .title h1 span { color: #FF6900; }
10
+ .title p { margin-top: 4px; color: #555; font-size: 1.05em; }
11
+ h3 { border-bottom: 1px solid #eee; padding-bottom: 6px; margin-top: 30px; }
12
+ table { width: 100%; border-collapse: collapse; margin: 15px 0; }
13
+ th, td { text-align: left; padding: 8px 12px; border: 1px solid #e0e0e0; }
14
+ th { background: #f7f7f7; font-weight: 600; }
15
+ a { color: #2563eb; text-decoration: none; }
16
+ a:hover { text-decoration: underline; }
17
+ ul { padding-left: 20px; }
18
+ li { margin: 6px 0; }
19
+ </style>
20
+ </head>
21
+ <body>
22
+
23
+ <div class="title">
24
+ <h1><span>UI-MOPD</span>: Multi-platform On-Policy Distillation for Continual GUI Agent Learning</h1>
25
+ </div>
26
+
27
+ <p>
28
+ We build cross-platform GUI agents that can operate both desktop and mobile interfaces through a unified training framework.
29
+ </p>
30
+
31
+ <h3>Research</h3>
32
+ <p>UI-MOPD introduces a two-stage training pipeline:</p>
33
+ <ul>
34
+ <li><b>Stage 1</b>: Supervised Fine-Tuning (SFT) on platform-specific teacher models</li>
35
+ <li><b>Stage 2</b>: Reinforcement Learning distillation (DAPO) with multi-teacher on-policy guidance</li>
36
+ </ul>
37
+ <p>Our student model (8B) learns from multiple 32B teacher models to achieve strong cross-platform GUI interaction capabilities. All models are based on <b>Qwen3-VL-Thinking</b> (thinking/reasoning variants).</p>
38
+
39
+ <h3>Models</h3>
40
+ <table>
41
+ <tr><th>Model</th><th>Size</th><th>Description</th></tr>
42
+ <tr><td><a href="https://huggingface.co/UI-MOPD/Qwen3-VL-32B-Thinking-Desktop-Teacher">Qwen3-VL-32B-Thinking-Desktop-Teacher</a></td><td>32B</td><td>Desktop platform teacher (thinking)</td></tr>
43
+ <tr><td><a href="https://huggingface.co/UI-MOPD/Qwen3-VL-32B-Thinking-Mobile-Teacher">Qwen3-VL-32B-Thinking-Mobile-Teacher</a></td><td>32B</td><td>Mobile platform teacher (thinking)</td></tr>
44
+ <tr><td><a href="https://huggingface.co/UI-MOPD/Qwen3-VL-8B-Thinking-Desktop-SFT">Qwen3-VL-8B-Thinking-Desktop-SFT</a></td><td>8B</td><td>Desktop SFT checkpoint (thinking)</td></tr>
45
+ <tr><td><a href="https://huggingface.co/UI-MOPD/Qwen3-VL-8B-Thinking-Mobile-SFT">Qwen3-VL-8B-Thinking-Mobile-SFT</a></td><td>8B</td><td>Mobile SFT checkpoint (thinking)</td></tr>
46
+ <tr><td><a href="https://huggingface.co/UI-MOPD/Qwen3-VL-8B-Thinking-UI-MOPD-Student">Qwen3-VL-8B-Thinking-UI-MOPD-Student</a></td><td>8B</td><td>Final cross-platform student (thinking)</td></tr>
47
+ </table>
48
+
49
+ <h3>Datasets</h3>
50
+ <table>
51
+ <tr><th>Dataset</th><th>Description</th></tr>
52
+ <tr><td><a href="https://huggingface.co/datasets/UI-MOPD/Uni-GUI-OpenCUA">Uni-GUI-OpenCUA</a></td><td>Post-processed desktop trajectories from OpenCUA (~832 episodes, ~14K steps)</td></tr>
53
+ <tr><td><a href="https://huggingface.co/datasets/UI-MOPD/Uni-GUI-Desktop-1">Uni-GUI-Desktop-1</a></td><td>Large-scale desktop GUI trajectories (~2.7K episodes, ~36K steps)</td></tr>
54
+ <tr><td><a href="https://huggingface.co/datasets/UI-MOPD/Uni-GUI-Desktop-2">Uni-GUI-Desktop-2</a></td><td>OSWorld desktop trajectories (~1.2K episodes, ~14.8K steps)</td></tr>
55
+ <tr><td><a href="https://huggingface.co/datasets/UI-MOPD/Uni-GUI-Mobile">Uni-GUI-Mobile</a></td><td>Mobile GUI trajectories (~871 episodes, ~14K steps)</td></tr>
56
+ <tr><td><a href="https://huggingface.co/datasets/UI-MOPD/Uni-GUI-OpenMobile">Uni-GUI-OpenMobile</a></td><td>Open-source Android app trajectories (~2.6K episodes, ~25.9K steps)</td></tr>
57
+ <tr><td><a href="https://huggingface.co/datasets/UI-MOPD/AndroidControl-Star">AndroidControl*</a></td><td>Static mobile GUI evaluation subset (4,260 step records, 781 episodes)</td></tr>
58
+ </table>
59
+
60
+ <h3>Evaluation Results</h3>
61
+ <table>
62
+ <tr><th>Benchmark</th><th>Description</th></tr>
63
+ <tr><td><a href="https://huggingface.co/datasets/UI-MOPD/OSWorld-Eval-Results">OSWorld-Eval-Results</a></td><td>Desktop evaluation on OSWorld (359 tasks, 35.1% success rate)</td></tr>
64
+ <tr><td><a href="https://huggingface.co/datasets/UI-MOPD/MobileWorld-Eval-Results">MobileWorld-Eval-Results</a></td><td>Mobile evaluation on MobileWorld (117 tasks, 10.3% success rate)</td></tr>
65
+ </table>
66
+
67
+ <h3>Links</h3>
68
+ <ul>
69
+ <li><a href="https://elispctre.github.io/UI-MOPD">Project Page</a></li>
70
+ </ul>
71
+
72
+ </body>
73
+ </html>