Datasets:
license: apache-2.0
task_categories:
- visual-question-answering
language:
- zh
pretty_name: CMMU
size_categories:
- 1K<n<10K
dataset_info:
features:
- name: type
dtype: string
- name: grade_band
dtype: string
- name: difficulty
dtype: string
- name: question_info
dtype: string
- name: split
dtype: string
- name: subject
dtype: string
- name: image
dtype: string
- name: sub_questions
sequence: string
- name: options
sequence: string
- name: answer
sequence: string
- name: solution_info
dtype: string
- name: id
dtype: string
- name: image
dtype: image
configs:
- config_name: default
data_files:
- split: val
path:
- val/*.parquet
CMMU
This repo contains the evaluation code for the paper CMMU: A Benchmark for Chinese Multi-modal Multi-type Question Understanding and Reasoning .
We release the validation set of CMMU, you can download it from here. The test set will be hosted on the flageval platform. Users can test by uploading their models.
Introduction
CMMU is a novel multi-modal benchmark designed to evaluate domain-specific knowledge across seven foundational subjects: math, biology, physics, chemistry, geography, politics, and history. It comprises 3603 questions, incorporating text and images, drawn from a range of Chinese exams. Spanning primary to high school levels, CMMU offers a thorough evaluation of model capabilities across different educational stages.
Evaluation Results
We currently evaluated 10 models on CMMU. The results are shown in the following table.
Model | Val Avg. | Test Avg. |
---|---|---|
InstructBLIP-13b | 0.39 | 0.48 |
CogVLM-7b | 5.55 | 4.9 |
ShareGPT4V-7b | 7.95 | 7.63 |
mPLUG-Owl2-7b | 8.69 | 8.58 |
LLava-1.5-13b | 11.36 | 11.96 |
Qwen-VL-Chat-7b | 11.71 | 12.14 |
Intern-XComposer-7b | 18.65 | 19.07 |
Gemini-Pro | 21.58 | 22.5 |
Qwen-VL-Plus | 26.77 | 26.9 |
GPT-4V | 30.19 | 30.91 |
Citation
BibTeX:
@article{he2024cmmu,
title={CMMU: A Benchmark for Chinese Multi-modal Multi-type Question Understanding and Reasoning},
author={Zheqi He, Xinya Wu, Pengfei Zhou, Richeng Xuan, Guang Liu, Xi Yang, Qiannan Zhu and Hua Huang},
journal={arXiv preprint arXiv:2401.14011},
year={2024},
}