Skip to content

Update README.md #23

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
QianhuiWu merged 1 commit into from
Jun 26, 2025
Merged

Update README.md #23

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 3 additions & 3 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -46,7 +46,7 @@ Figure 1. **Left**: Model performance vs. training data scale on the ScreenSpot

💡 **Rethink how humans interact with digital interfaces**: humans do NOT calculate precise screen coordinates before acting—they perceive the target element and interact with it directly.

🚀 **We propose _GUI-Actor_, a VLM enhanced by an action head, to mitigate the above limitations.** The attention-based action head not only enables GUI-Actor to peform coordinate-free GUI grounding that more closely aligns with human behavior, but also can generate multiple candidate regions in a single forward pass, offering flexibility for downstream modules such as search strategies.
🚀 **We propose _GUI-Actor_, a VLM enhanced by an action head, to mitigate the above limitations.** The attention-based action head not only enables GUI-Actor to perform coordinate-free GUI grounding that more closely aligns with human behavior, but also can generate multiple candidate regions in a single forward pass, offering flexibility for downstream modules such as search strategies.

➕ **We design a _grounding verifier_ to evaluate and select the most plausible action region** among the candidates proposed from the action attention map. We show that this verifier can be easily integrated with other grounding methods for a further performance boost.

Expand Down Expand Up @@ -161,7 +161,7 @@ model = Qwen2VLForConditionalGenerationWithPointer.from_pretrained(
# prepare example
dataset = load_dataset("rootsautomation/ScreenSpot")["test"]
example = dataset[0]
print(f"Intruction: {example['instruction']}")
print(f"Instruction: {example['instruction']}")
print(f"ground-truth action region (x1, y1, x2, y2): {[round(i, 2) for i in example['bbox']]}")

conversation = [
Expand Down Expand Up @@ -196,7 +196,7 @@ px, py = pred["topk_points"][0]
print(f"Predicted click point: [{round(px, 4)}, {round(py, 4)}]")

# >> Model Response
# Intruction: close this window
# Instruction: close this window
# ground-truth action region (x1, y1, x2, y2): [0.9479, 0.1444, 0.9938, 0.2074]
# Predicted click point: [0.9709, 0.1548]
```
Expand Down