VLMEvalKit
收藏arXiv2025-09-30 收录
下载链接:
https://huggingface.co/datasets/VLMEval/OpenVLMRecords
下载链接
链接失效反馈官方服务:
资源简介:
该数据集汇总了数百个多模态大型语言模型(MLLMs)在各个基准测试中的评估结果。它包含了来自18个公开可获取基准测试的评估数据,并支持对MLLM评估中冗余性的分析。该数据集的任务是对多模态大型语言模型(MLLMs)的基准评估。
This dataset aggregates evaluation results of hundreds of multimodal large language models (MLLMs) across various benchmark tests. It includes evaluation data from 18 publicly accessible benchmark tests, and supports analysis of redundancy in MLLM evaluations. This dataset is intended for benchmark evaluations of multimodal large language models (MLLMs).
提供机构:
HuggingFace



