|
|
|
URL = "http://opencompass.openxlab.space/assets/MathLB.json" |
|
|
|
CITATION_BUTTON_TEXT = r"""\ |
|
@inproceedings{duan2024vlmevalkit, |
|
title={Vlmevalkit: An open-source toolkit for evaluating large multi-modality models}, |
|
author={Duan, Haodong and Yang, Junming and Qiao, Yuxuan and Fang, Xinyu and Chen, Lin and Liu, Yuan and Dong, Xiaoyi and Zang, Yuhang and Zhang, Pan and Wang, Jiaqi and others}, |
|
booktitle={Proceedings of the 32nd ACM International Conference on Multimedia}, |
|
pages={11198--11201}, |
|
year={2024} |
|
} |
|
""" |
|
CITATION_BUTTON_LABEL = "Copy the following snippet to cite these results" |
|
|
|
LEADERBORAD_INTRODUCTION = """# Open LMM Reasoning Leaderboard |
|
|
|
This leaderboard aims at providing a comprehensive evaluation of the reasoning capabilities of LMMs. \ |
|
Currently, it is a collection of evaluation results on multiple multi-modal mathematical reasoning benchmarks. |
|
""" |
|
|
|
|
|
DATASETS_ALL = ['MathVista', 'MathVision', 'MathVerse', 'DynaMath', 'MMMath', 'OlympiadBench'] |
|
DATASETS_ESS = ['MathVista', 'MathVision', 'MathVerse', 'DynaMath'] |
|
META_FIELDS = ['Method', 'Param (B)', 'Language Model', 'Vision Model', 'OpenSource', 'Verified', 'Org'] |
|
MODEL_SIZE = ['<4B', '4B-10B', '10B-20B', '20B-40B', '>40B', 'Unknown'] |
|
MODEL_TYPE = ['OpenSource', 'API'] |
|
|
