Yuxiang007/PIRA-Bench-data

Dataset

PIRA-Bench

0

7 commits

1 linked in READMEs

updated Mar 16, 2026

See the code

README

PIRA-Bench

Trajectory data for PIRA-Bench: A Transition from Reactive GUI Agents to GUI-based Proactive Intent Recommendation Agents.

Project Page: https://www.pira-bench.top

Unzip trajs.zip and it contains 100 trajectory folders where each folder contains several images and a profile.json. For each profile, you need to predict the intents and you need to finalize a json file in the following format:

{
  "traj_1": {
    "Profile_A": ["intent 1", "intent 2"],
    "Profile_B": ["intent 3"],
    "Profile_C": []
  },
  "traj_2": ...
}

Then you can upload this inference result to the leaderboard for evaluation and submission.

Contributors

Yuxiang007

6 commits

nielsr

1 commits

Yuxiang007/PIRA-Bench-data

Dataset

PIRA-Bench

0

7 commits

1 linked in READMEs

updated Mar 16, 2026

See the code

README

PIRA-Bench

Trajectory data for PIRA-Bench: A Transition from Reactive GUI Agents to GUI-based Proactive Intent Recommendation Agents.

Project Page: https://www.pira-bench.top

Unzip trajs.zip and it contains 100 trajectory folders where each folder contains several images and a profile.json. For each profile, you need to predict the intents and you need to finalize a json file in the following format:

{
  "traj_1": {
    "Profile_A": ["intent 1", "intent 2"],
    "Profile_B": ["intent 3"],
    "Profile_C": []
  },
  "traj_2": ...
}

Then you can upload this inference result to the leaderboard for evaluation and submission.

Contributors

Yuxiang007

6 commits

nielsr

1 commits