Update README.md
Browse files
README.md
CHANGED
|
@@ -10,6 +10,7 @@ tags:
|
|
| 10 |
- art
|
| 11 |
- text-rendering
|
| 12 |
---
|
|
|
|
| 13 |
|
| 14 |
Use this code for inference:
|
| 15 |
```python
|
|
|
|
| 10 |
- art
|
| 11 |
- text-rendering
|
| 12 |
---
|
| 13 |
+
**LeX-Enhancer** is a lightweight prompt enhancement model distilled from DeepSeek-R1. Specifically, we collect 60,856 prompt pairs before and after R1 enhancement, termed as LeX-R1-60K, and fine-tune a Qwen2.5-14B model using LoRA to replicate the detailed prompting capabilities of R1. This enables efficient, large-scale generation of high-quality, visually grounded prompts.
|
| 14 |
|
| 15 |
Use this code for inference:
|
| 16 |
```python
|