ViLBench

What is ViLBench?

ViLBench is a benchmark designed to evaluate vision-language models. It features 600 examples from 5 datasets, selected based on the criterion that process…

Released

2025-03-26

Modality

multimodal

Evaluates

多模态, Multimodal, 推理, Reasoning, vision-language, math, 多模态模型, VLM, 逻辑推理, 不支持, Unsupported

Openness

unknown

Importer

OpenCompass Hub

Reported scores

0

Source provenance

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

ViLBench paper and dataset

• Paper https://arxiv.org/abs/2503.20271
• Dataset https://huggingface.co/datasets/UCSC-VLAA/ViLBench

Which sources cite ViLBench?

• ViLBench

Related 多模态, Multimodal, 推理, Reasoning, vision-language, math, 多模态模型, VLM, 逻辑推理, 不支持, Unsupported benchmarks

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