highlight verifier numbers.
Browse files
README.md
CHANGED
@@ -60,8 +60,8 @@ Table 2. Main results on the ScreenSpot-Pro and ScreenSpot-v2 with **Qwen2.5-VL*
|
|
60 |
| **_3B models:_**
|
61 |
| Qwen2.5-VL-3B | Qwen2.5-VL | 25.9 | 80.9 |
|
62 |
| Jedi-3B | Qwen2.5-VL | 36.1 | 88.6 |
|
63 |
-
| GUI-Actor-3B | Qwen2.5-VL |
|
64 |
-
| GUI-Actor-3B + Verifier | Qwen2.5-VL | 45.9 | 92.4 |
|
65 |
|
66 |
## 🚀 Usage
|
67 |
The verifier takes a language instruction and an image with a red circle marking the target position as input. One example is shown below. It outputs either ‘True’ or ‘False’, and you can also use the probability of each label to score the sample.
|
|
|
60 |
| **_3B models:_**
|
61 |
| Qwen2.5-VL-3B | Qwen2.5-VL | 25.9 | 80.9 |
|
62 |
| Jedi-3B | Qwen2.5-VL | 36.1 | 88.6 |
|
63 |
+
| GUI-Actor-3B | Qwen2.5-VL | 42.2 | 91.0 |
|
64 |
+
| GUI-Actor-3B + Verifier | Qwen2.5-VL | **45.9** | **92.4** |
|
65 |
|
66 |
## 🚀 Usage
|
67 |
The verifier takes a language instruction and an image with a red circle marking the target position as input. One example is shown below. It outputs either ‘True’ or ‘False’, and you can also use the probability of each label to score the sample.
|