A-Bench

What is A-Bench?

A-Bench is a benchmark designed to diagnose whether LMMs are masters at evaluating AIGIs. 2,864 AIGIs from 16 text-to-image models are sampled, each paired…

Released

2024-06-05

Modality

multimodal

Evaluates

多模态, Multimodal, 多模态模型, VLM, 视觉生成, Visual Generation, 图像理解, Image Understanding, 不支持, Unsupported

Openness

unknown

Importer

OpenCompass Hub

Reported scores

0

Source provenance

• Original evidence https://hub.opencompass.org.cn/dataset-detail/A-Bench

A-Bench paper, code and dataset

• Paper https://arxiv.org/abs/2406.03070
• Code https://github.com/Q-Future/A-Bench
• Dataset https://huggingface.co/datasets/q-future/A-Bench

Which sources cite A-Bench?

• A-Bench

Related 多模态, Multimodal, 多模态模型, VLM, 视觉生成, Visual Generation, 图像理解, Image Understanding, 不支持, Unsupported benchmarks

• 3DCodeBench
• A-OKVQA
• AbsPyramid
• ActionAtlas
• ActiveFly-Bench
• Ada-LEval
• AECBench
• AFQMC