Update README.md

README.md

@@ -26,14 +26,11 @@ Our model is composed of two components: the [Qwen2.5-VL-3B-Instruct](https://hu
 
 ![Model](./assets/model.png)
 
-## Performance
 
-### Qualitative Results for Text-to-Image Task
-![t2i](./assets/t2i.png)
+## Text-to-Image Generation Performance
 
-### Qualitative Results for Image-Text-to-Image Task
-![t2i](./assets/ti2i.png)
-**The input image is processed by the Qwen2.5-VL image encoder and then fed into the MLLM along with text and learnable queries. We use only the learnable queries, which have fused the multimodal information, as the generative condition, without directly incorporating any image VAE representations like other works. Despite this, the model still achieves promising multimodal understanding and consistency performance in Image-Text-to-Image tasks.**
+### Qualitative Results
+![t2i](./assets/t2i.png)
 
 ### GenEval and DPG-Bench
 | Method | Base (M)LLM | GenEval | DPG-Bench |
@@ -54,6 +51,13 @@ Our model is composed of two components: the [Qwen2.5-VL-3B-Instruct](https://hu
 
 ![TIIF](./assets/tiif_bench.png)
 
+## Image Editing Performance
+
+The input image is processed by the Qwen2.5-VL image encoder and then fed into the MLLM along with text and learnable queries. We use only the learnable queries, which have fused the multimodal information, as the generative condition, without directly incorporating any image VAE representations like other works. Despite this, the model still achieves promising multimodal understanding and consistency performance in Image Editing tasks.
+
+### Qualitative Results
+![t2i](./assets/ti2i.png)
+
 ### ImgEdit
 
 ![ImgEdit](./assets/imgedit.png)