MM-Hallu/MMMC
收藏Hugging Face2026-04-25 更新2026-05-03 收录
下载链接：
https://hf-mirror.com/datasets/MM-Hallu/MMMC
下载链接
链接失效反馈官方服务：
资源简介：
---
dataset_info:
features:
- name: image
dtype: image
- name: image_id
dtype: string
- name: question
dtype: string
- name: answer
dtype: string
- name: conflict_type
dtype: string
- name: key_component
dtype: string
- name: key_component_relationships
dtype: string
- name: key_component_attributes
dtype: string
configs:
- config_name: default
data_files:
- split: train
path: train-*.parquet
- split: test
path: test-*.parquet
license: apache-2.0
task_categories:
- visual-question-answering
language:
- en
tags:
- multimodal
- conflict
- hallucination
- benchmark
size_categories:
- 10K<n<100K
---
# MMMC
Multimodal Modality Conflict benchmark with 40,000 examples. Tests VLMs' ability to handle conflicting information between image and text modalities.
## Fields
| Field | Description |
|-------|-------------|
| image | Input image |
| image_id | Image identifier |
| question | Question about the image |
| answer | Ground truth answer |
| conflict_type | Type of modality conflict |
| key_component | JSON-encoded key components |
| key_component_relationships | JSON-encoded component relationships |
| key_component_attributes | JSON-encoded component attributes |
Source: [ustc-zhangzm/MMMC](https://huggingface.co/datasets/ustc-zhangzm/MMMC)
提供机构：
MM-Hallu
