datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
Typographic-Dataset
Unveiling Typographic Deceptions: Insights of the Typographic Vulnerability in Large Vision-Language Model
Hao Cheng*,
Erjia Xiao*,
Jindong Gu,
Le Yang,
Jinhao Duan,
Jize Zhang,
Jiahang Cao,
Kaidi Xu,
Renjing Xu†
HKUST & University of Oxford & Drexel University & Xi’an Jiaotong University
Introduction
The Typographic Dataset is curated to explore the impact of… See the full description on the dataset page: https://huggingface.co/datasets/erjiaxiao/Typographic-Dataset.Typographic-Visual-Prompt-Injection-Dataset
Exploring Typographic Visual Prompts Injection Threats in Cross-Modality Generation Models
Hao Cheng*, Erjia Xiao*, Yichi Wang, Lingfeng Zhang, Qiang Zhang, Jiahang Cao, Kaidi Xu, Mengshu Sun
Xiaoshuai Hao†, Jindong Gu†, Renjing Xu†
The Hong Kong University of Science and Technology (Guangzhou) & University of Oxford Beijing Academy of Artificial Intelligence & Beijing University of Technology Tsinghua University & Drexel University & X-Humanoid… See the full description on the dataset page: https://huggingface.co/datasets/erjiaxiao/Typographic-Visual-Prompt-Injection-Dataset.CLIP-adversarial-typographic-attack_text-image
CLIP-adversarial-typographic-attack_text-image
A typographic attack dataset for CLIP. For adversarial training & model research / XAI (research) use.
First 47 are random and self-made images, rest are from dataset: SPRIGHT-T2I/spright_coco. Of which:
Images are selected for pre-trained OpenAI/CLIP ViT-L/14 features; for highly salient 'text related' concepts via Sparse Autoencoder (SAE).
Labels via CLIP ViT-L/14 gradient ascent -> optimize text embeddings for cosine… See the full description on the dataset page: https://huggingface.co/datasets/zer0int/CLIP-adversarial-typographic-attack_text-image.40_typographic_error_data_v1.3
Dataset Card for "40_typographic_error_data_v1.3"
More Information needed
safety_typographic_jailbreaking