Datasets:
You need to agree to share your contact information to access this dataset
This repository is publicly accessible, but you have to accept the conditions to access its files and content.
Access to the Re:Verse Dataset is granted for non-commercial research purposes only. Please ensure all information is accurate and complete your request below.
Log in or Sign Up to review the conditions and access this dataset content.
Re:Verse - Can Your VLM Read a Manga?
📄 1st AI Story Workshop ICCV (Oral)
- Paper: https://arxiv.org/abs/2508.08508
- Website: https://re-verse.verel.app
Overview
Re:Verse introduces a comprehensive benchmark for evaluating Vision Language Models (VLMs) on manga reading comprehension. Our framework tests models' ability to understand sequential visual narratives, character relationships, and temporal reasoning across manga panels.
Dataset
The Re:Verse dataset contains:
- Multi-panel manga sequences from diverse genres
- Structured annotations for story comprehension tasks
Citation
@inproceedings{
baranwal2025reverse,
title={Re:Verse - Can your VLM read a Manga?},
author={Aaditya Baranwal and Madhav Kataria and Naitik Agrawal and Shruti Vyas and Yogesh S Rawat},
booktitle={Proceedings of the IEEE/CVF International Conference on Computer Vision - Workshops (ICCV-W)},
year={2025},
url={https://arxiv.org/abs/2508.08508}
}
Authors
Aaditya Baranwal¹†, Madhav Kataria², Naitik Agrawal³, Yogesh Singh Rawat¹, Shruti Vyas¹
¹University of Central Florida || ²Indian Institute of Technology Jodhpur || ³Indian Institute of Technology Varanasi
- Downloads last month
- 6