princeton-nlp/CharXiv
Visual Question AnsweringDocument Question AnsweringQuestion AnsweringEN
The princeton-nlp/CharXiv dataset is a EN visual question answering resource from princeton-nlp at 2024.
About princeton-nlp/CharXiv
CharXiv: Charting Gaps in Realistic Chart Understanding in Multimodal LLMs
NeurIPS 2024
š Home (š§Still in construction) | š¤Data | š„Leaderboard | š„ļøCode | šPaper
This repo contains the full dataset for our paper CharXiv: Charting Gaps in Realisti...
Details
- Task
- Visual Question Answering, Document Question Answering, Question Answering
- Language
- EN
- Format
- Parquet
- Rows / instances
- N/A
- Creator
- princeton-nlp
- Year
- 2024