datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
GUIGuard-Bench
GUIGuard-Bench (Public Ladder)
GUIGuard-Bench is a cross-platform GUI agent benchmark for studying privacy risks and privacy-preserving execution in multimodal GUI agents.
This public-ladder release contains 121 GUI interaction trajectories (68 Android + 53 PC) for benchmark evaluation, with 26,407 region-level privacy annotations across 2,002 screenshots.
For the anonymous review version of the evaluation toolkit, see GUIGaurd-Bench-CA4F.
Dataset Summary
GUI agents… See the full description on the dataset page: https://huggingface.co/datasets/ShaofantuoshuzhengzhiSha/GUIGuard-Bench.two-box-judge-gui
Two-Box Judge GUI Dataset
A multimodal dataset for training GUI element selection models. Given two candidate bounding boxes on a GUI screenshot, the model learns to select the one that better fulfills the user's intent.
Dataset Description
This dataset is designed for training judge models in GUI grounding pipelines. When a visual grounding model produces multiple candidate regions, the judge model determines which candidate best matches the user's command.… See the full description on the dataset page: https://huggingface.co/datasets/THU-BoZhang/two-box-judge-gui.reef-guidance-system
Dataset Card for Reef Guidance System
This dataset provides imagery used for training and evaluation of models in the Reef Guidance System. All imagery was collected by the Australian Institute of Marine Science using the ReefScan™ Transom Marine Monitoring System.
If you use this dataset in your work, please cite the associated paper: AI-driven dispensing of coral reseeding devices for broad-scale restoration of the Great Barrier Reef (citations provided at bottom of this… See the full description on the dataset page: https://huggingface.co/datasets/QCR-Underwater-Perception/reef-guidance-system.gui-odyssey-train
Dataset Card for GUI Odyssey (Train Split)
⬆️ Test split shown above, but this also represents the train split.
This is a FiftyOne dataset with 89365 samples.
Installation
If you haven't already, install FiftyOne:
pip install -U fiftyone
Usage
import fiftyone as fo
from fiftyone.utils.huggingface import load_from_hub
# Load the dataset
# Note: other available arguments include 'max_samples', etc
dataset = load_from_hub("Voxel51/gui-odyssey-train")
# Launch… See the full description on the dataset page: https://huggingface.co/datasets/Voxel51/gui-odyssey-train.gui-odyssey-test
Dataset Card for GUI Odyssey (Test Split)
This is a FiftyOne dataset with 29426 samples.
Installation
If you haven't already, install FiftyOne:
pip install -U fiftyone
Usage
import fiftyone as fo
from fiftyone.utils.huggingface import load_from_hub
# Load the dataset
# Note: other available arguments include 'max_samples', etc
dataset = load_from_hub("Voxel51/gui-odyssey-test")
# Launch the App
session = fo.launch_app(dataset)
Dataset Details… See the full description on the dataset page: https://huggingface.co/datasets/Voxel51/gui-odyssey-test.guiact_websingle_test
Dataset Card for GUIAct Web-Single Dataset - Test Set
This is a FiftyOne dataset with 1410 samples.
Installation
If you haven't already, install FiftyOne:
pip install -U fiftyone
Usage
import fiftyone as fo
from fiftyone.utils.huggingface import load_from_hub
# Load the dataset
# Note: other available arguments include 'max_samples', etc
dataset = load_from_hub("Voxel51/guiact_websingle_test")
# Launch the App
session = fo.launch_app(dataset)… See the full description on the dataset page: https://huggingface.co/datasets/Voxel51/guiact_websingle_test.low-guidance-cfg-sweep
Sub-CFG guidance sweep (g = 0 → 2), SDXL + SD 3.5
Exploratory. Not pre-registered. Not a result.
No hypothesis was committed before these runs, there is no pre-specified
statistical model, and no p-values are reported anywhere in this dataset.
The sibling Exp 03 dataset
is pre-registered, with commit dates as proof. This one is not. Treat it
as a reason to design an experiment, not as evidence for a claim.
From the Operating System Hypothesis project. Exp 01 and Exp 03 both… See the full description on the dataset page: https://huggingface.co/datasets/youssefhassan13/low-guidance-cfg-sweep.two-box-judge-gui-sharded
Two-Box Judge GUI Dataset (Sharded)
A multimodal dataset for training GUI element selection models, packaged in WebDataset format for efficient streaming.
Dataset Statistics
Split
Samples
Shards
Size
Train
115,638
6
25.32 GB
Validation
12,849
1
2.82 GB
Format
This dataset uses WebDataset format - sharded tar.gz archives for efficient streaming:
train/
├── shard-00000.tar.gz
├── shard-00001.tar.gz
└── ...
Each shard contains… See the full description on the dataset page: https://huggingface.co/datasets/THU-BoZhang/two-box-judge-gui-sharded.screenspot_pro_gui_actor
Dataset Card for Voxel51/ScreenSpot-Pro
This is a FiftyOne dataset with 1581 samples.
Installation
If you haven't already, install FiftyOne:
pip install -U fiftyone
Usage
import fiftyone as fo
from fiftyone.utils.huggingface import load_from_hub
# Load the dataset
# Note: other available arguments include 'max_samples', etc
dataset = load_from_hub("harpreetsahota/screenspot_pro_gui_actor")
# Launch the App
session = fo.launch_app(dataset)… See the full description on the dataset page: https://huggingface.co/datasets/harpreetsahota/screenspot_pro_gui_actor.human-activity-pose_v4
🧍 Human Activity Pose Dataset (Split Version)
This dataset contains human pose landmarks extracted with MediaPipe Pose,
annotated with activity labels and textual descriptions in English.
Dataset structure
train/ — 80% of samples for training
validation/ — 20% of samples for validation
Each record includes:
33 pose keypoints (fields: x, y, z, visibility)
label: activity name (e.g., reading, dancing, office_work)
description: a short textual description of the action… See the full description on the dataset page: https://huggingface.co/datasets/guillherms/human-activity-pose_v4.autotrain-data-guitarsproject
AutoTrain Dataset for project: guitarsproject
Dataset Description
This dataset has been automatically processed by AutoTrain for project guitarsproject.
Languages
The BCP-47 code for the dataset's language is unk.
Dataset Structure
Data Instances
A sample from this dataset looks as follows:
[
{
"image": "<1990x2520 RGB PIL image>",
"target": 1
},
{
"image": "<6000x4000 RGB PIL image>",
"target": 0
}
]… See the full description on the dataset page: https://huggingface.co/datasets/rpereira90/autotrain-data-guitarsproject.screenspot_v2_w_gui_actor
Dataset Card for Voxel51/ScreenSpot-v2
This is a FiftyOne dataset with 1272 samples.
Installation
If you haven't already, install FiftyOne:
pip install -U fiftyone
Usage
import fiftyone as fo
from fiftyone.utils.huggingface import load_from_hub
# Load the dataset
# Note: other available arguments include 'max_samples', etc
dataset = load_from_hub("harpreetsahota/screenspot_v2_w_gui_actor")
# Launch the App
session = fo.launch_app(dataset)… See the full description on the dataset page: https://huggingface.co/datasets/harpreetsahota/screenspot_v2_w_gui_actor.guiact_websingle_test
Dataset Card for GUIAct Web-Single Dataset - Test Set
This is a FiftyOne dataset with 1410 samples.
Installation
If you haven't already, install FiftyOne:
pip install -U fiftyone
Usage
import fiftyone as fo
from fiftyone.utils.huggingface import load_from_hub
# Load the dataset
# Note: other available arguments include 'max_samples', etc
dataset = load_from_hub("Voxel51/guiact_websingle_test")
# Launch the App
session = fo.launch_app(dataset)… See the full description on the dataset page: https://huggingface.co/datasets/ZhuOnR/guiact_websingle_test.