metadata
title: README
emoji: 🖥️
colorFrom: blue
colorTo: purple
sdk: static
pinned: false
🖥️📱 UI-MOPD
Multi-platform On-Policy Distillation for Continual GUI Agent Learning
We build cross-platform GUI agents that can operate both desktop and mobile interfaces through a unified training framework.
🔬 Research
UI-MOPD introduces a two-stage training pipeline:
- Stage 1: Supervised Fine-Tuning (SFT) on platform-specific teacher models
- Stage 2: Reinforcement Learning distillation (DAPO) with multi-teacher on-policy guidance
Our student model (8B) learns from multiple 32B teacher models to achieve strong cross-platform GUI interaction capabilities. All models are based on Qwen3-VL-Thinking (thinking/reasoning variants).
🤖 Models
| Model | Size | Description |
|---|---|---|
| Qwen3-VL-32B-Thinking-Desktop-Teacher | 32B | Desktop platform teacher (thinking) |
| Qwen3-VL-32B-Thinking-Mobile-Teacher | 32B | Mobile platform teacher (thinking) |
| Qwen3-VL-8B-Thinking-Desktop-SFT | 8B | Desktop SFT checkpoint (thinking) |
| Qwen3-VL-8B-Thinking-Mobile-SFT | 8B | Mobile SFT checkpoint (thinking) |
| Qwen3-VL-8B-Thinking-UI-MOPD-Student | 8B | Final cross-platform student (thinking) |
📊 Datasets
| Dataset | Description |
|---|---|
| Uni-GUI-OpenCUA | Post-processed desktop trajectories from OpenCUA (~832 episodes, ~14K steps) |
| Uni-GUI-Desktop-1 | Large-scale desktop GUI trajectories (~2.7K episodes, ~36K steps) |
| Uni-GUI-Desktop-2 | OSWorld desktop trajectories (~1.2K episodes, ~14.8K steps) |
| Uni-GUI-Mobile | Mobile GUI trajectories (~871 episodes, ~14K steps) |
| Uni-GUI-OpenMobile | Open-source Android app trajectories (~2.6K episodes, ~25.9K steps) |
| AndroidControl* | Static mobile GUI evaluation subset (4,260 step records, 781 episodes) |
📈 Evaluation Results
| Benchmark | Description |
|---|---|
| OSWorld-Eval-Results | Desktop evaluation on OSWorld (359 tasks, 35.1% success rate) |
| MobileWorld-Eval-Results | Mobile evaluation on MobileWorld (117 tasks, 10.3% success rate) |