update

app.py

@@ -6,7 +6,7 @@ from transformers import AutoProcessor, Qwen2_5_VLForConditionalGeneration
 
 MODEL_ID = "internlm/CapRL-3B"
 DEFAULT_PROMPT = "Describe the image in detail."
-MAX_NEW_TOKENS = 1024
+MAX_NEW_TOKENS = 4096
 
 
 def get_device() -> str:
@@ -124,13 +124,26 @@ with gr.Blocks(title="CapRL Image Captioning") as demo:
         outputs=[caption_output, token_output],
     )
 
+    gr.Examples(
+        examples=[
+            ["./examples/example_chinese.png"],
+            ["./examples/example_receipt.jpg"],
+            ["./examples/example_table.png"],
+        ],
+        inputs=image_input,
+        outputs=[caption_output, token_output],
+        fn=generate_caption,
+        cache_examples=False,
+        label="📸 Example Images"
+    )
+
     gr.Markdown(
         """### Citation
 If you find this project useful, please kindly cite:
 
 ```
 @article{xing2025caprl,
-  title={CapRL: Stimulating Dense Image Caption Capabilities via Reinforcement Learning},
+  title={{CapRL}: Stimulating Dense Image Caption Capabilities via Reinforcement Learning},
   author={Xing, Long and Dong, Xiaoyi and Zang, Yuhang and Cao, Yuhang and Liang, Jianze and Huang, Qidong and Wang, Jiaqi and Wu, Feng and Lin, Dahua},
   journal={arXiv preprint arXiv:2509.22647},
   year={2025}