Update README.md
Browse files
README.md
CHANGED
|
@@ -38,7 +38,7 @@ QwenGuard-v1.2-3B is trained on [LlavaGuard-DS](https://huggingface.co/datasets/
|
|
| 38 |
|
| 39 |
|
| 40 |
## Overview
|
| 41 |
-
We here provide the
|
| 42 |
It builds upon Qwen/Qwen2.5-VL-3B-Instruct and shows improved reasoning capabilities within the rationales.
|
| 43 |
|
| 44 |
|
|
|
|
| 38 |
|
| 39 |
|
| 40 |
## Overview
|
| 41 |
+
We here provide the transformer weights for QwenGuard v1.2 3B. QwenGuard is a vision safeguard model that assesses images based on a provided safety policy, providing safety rating, safety category and a rationale.
|
| 42 |
It builds upon Qwen/Qwen2.5-VL-3B-Instruct and shows improved reasoning capabilities within the rationales.
|
| 43 |
|
| 44 |
|