Imaging · dataset · 2026
AdvSpot
Listed in Hugging Face Datasets
AdvSpot AdvSpot is the first grounded adversarial OCR benchmark, introduced in the paper ArmorOCR: Grounded Adversarial Visual Perception via Observation-Transferred Self-Distillation.
Description
It targets visual text patterns that are readable to humans but challenging for multimodal large language models (MLLMs) to localize and recognize, and evaluates models with region-level annotations: bounding boxes, transcriptions, perception-type labels, and region-grounded VQA pairs.
This… See the full description on the dataset page: huggingface.co/datasets/inclusionAI/advspot-public.
Links
Where it is published
- Hugging Face dataset page huggingface.co/datasets/inclusionAI/advspot-public ↗
landing page · from Hugging Face
Documentation and papers
- arXiv:2608.20122 arxiv.org/abs/2608.20122 ↗
publication · from Hugging Face
Catalogue records · 1
- Hub API huggingface.co/api/datasets/inclusionAI/advspot-public ↗
metadata API · from Hugging Face
Topics
- Stated by source
- image · image to text · text · visual question answering
- From keywords
- Computer Science & AI · Optical character recognition
- Inferred from text
- Machine learning 72% · Text 75%
Provenance · 1 source records, 15 field assertions
| Source | Key | Last seen | Raw |
|---|---|---|---|
| Hugging Face Datasets | inclusionAI/advspot-public | 8 d ago | JSON v1 |
| Field | Assertion | Extractor | Evidence |
|---|---|---|---|
| access_level | source · Hugging Face | connector:huggingface@1.0.0 | /gated |
| concepts[field].anzsrc:group:4611 | enrichment · Hugging Face | taxonomy-embedding@1.1.0 | title+keywords+description (72%) |
| concepts[field].local:field:computer-science-ai | mapping · Hugging Face | connector:huggingface@1.0.0 | |
| concepts[method].local:method:ocr | mapping · Hugging Face | vocabulary-mapper@1.0.0 | keywords['ocr'] |
| concepts[modality].hf_modality:image | source · Hugging Face | connector:huggingface@1.0.0 | |
| concepts[modality].hf_modality:text | source · Hugging Face | connector:huggingface@1.0.0 | |
| concepts[modality].local:modality:text | enrichment · Hugging Face | keyword-concept-rules@1.0.0 | title+description (75%) |
| concepts[task].hf_task:image-to-text | source · Hugging Face | connector:huggingface@1.0.0 | /tags[task_categories:*] |
| concepts[task].hf_task:visual-question-answering | source · Hugging Face | connector:huggingface@1.0.0 | /tags[task_categories:*] |
| created_date | source · Hugging Face | connector:huggingface@1.0.0 | |
| description | source · Hugging Face | connector:huggingface@1.0.0 | /description |
| license | source · Hugging Face | connector:huggingface@1.0.0 | /tags[license:*] |
| publication_date | source · Hugging Face | connector:huggingface@1.0.0 | |
| title | source · Hugging Face | connector:huggingface@1.0.0 | /id |
| updated_date | source · Hugging Face | connector:huggingface@1.0.0 |