init

llava/conversation.py

@@ -196,7 +196,8 @@ class Conversation:
                     img_str = f'<img src="data:image/jpeg;base64,{img_b64_str}" alt="user upload image" />'
                     msg=img_str
                     if extra_coordinates:
-                        ret.append(["The element is at {extra_coordinates} on the screen:",None])
+                        # msg=ret.append(["The element is at {extra_coordinates} on the screen:",None])+msg
+                        msg = f"The element is at {extra_coordinates} on the screen:"+ msg
                     ret.append([msg, None])
                 break
         return ret

llava/serve/gradio_web_server.py

@@ -397,7 +397,7 @@ def http_bot(state, model_selector, temperature, top_p, max_new_tokens, request:
 
 title_markdown = ("""
 # UGround: Universal Visual Grounding for GUI Agents
-[[Project Homepage](https://osu-nlp-group.github.io/UGround/)] [[Code](https://github.com/OSU-NLP-Group/UGround)] [[Model](https://huggingface.co/osunlp/UGround)] | 📚 [[Paper](https://arxiv.org/abs/2410.05243)]]
+[[🏠Project Homepage](https://osu-nlp-group.github.io/UGround/)] [[Code](https://github.com/OSU-NLP-Group/UGround)] [[😊Model](https://huggingface.co/osunlp/UGround)][[📚Paper](https://arxiv.org/abs/2410.05243)]
 """)
 
 tos_markdown = ("""
@@ -466,7 +466,7 @@ def build_demo(embed_mode, cur_dir=None, concurrency_count=1):
                 # max_output_tokens=16384
                 #
                 with gr.Accordion("Parameters", open=False) as parameter_row:
-                    temperature = gr.Slider(minimum=0.0, maximum=1.0, value=0.1, step=0.1, interactive=True, label="Temperature",)
+                    temperature = gr.Slider(minimum=0.0, maximum=1.0, value=0, step=0.1, interactive=True, label="Temperature",)
                     top_p = gr.Slider(minimum=0.0, maximum=1.0, value=0, step=0.1, interactive=True, label="Top P",)
                     max_output_tokens = gr.Slider(minimum=0, maximum=1024, value=512, step=64, interactive=True, label="Max output tokens",)
 

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "uground_demo_test"
-version = "4.1"
+version = "4.2"
 description = "Navigating the Digital World as Humans Do: Universal Visual Grounding for GUI Agents"
 readme = "README.md"
 requires-python = ">=3.8"