datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
Matterport3D-ScansAEGISThis repository contains the data of the paper [AEGIS: A Holistic Benchmark for Evaluating Forensic Analysis of AI-Generated Academic Images]
imageAEGIS-baseline
🎯 AEGIS-Baseline Dataset
AEGIS-Baseline is a practical subset of the AEGIS benchmark that contains all samples from the training and validation splits.
It is specifically designed for quick deployment and baseline method reproduction, offering lightweight, ready-to-use inputs and ground-truth labels.
The home page of AEGIS please refer to AEGIS
For complete access to the full dataset including raw videos, features, and training splits, please refer to AEGIS-Full.
📦… See the full description on the dataset page: https://huggingface.co/datasets/Clarifiedfish/AEGIS-baseline.AEGIS
🎬 AEGIS
AEGIS (Authenticity Evaluation on Generated vIdeo Samples) is a challenging benchmark that rigorously evaluates multimodal models’ capability in AI-generated video detection and reasoning.
This page hosts the AEGIS Hard Test Set, a carefully curated subset designed to stress-test model robustness under realistic, deceptive scenarios.
For complete access to the full dataset including raw videos, features, and training splits, please refer to AEGIS-Full.
🔗 Related… See the full description on the dataset page: https://huggingface.co/datasets/Clarifiedfish/AEGIS.balinese-carving-captionsimnet1k_breastplate_aegis_egis