HallusionBench

What is HallusionBench?

HallusionBench is a comprehensive benchmark designed for the evaluation of image-context reasoning, comprising 346 images paired with 1129 questions.

Released

2023-10-23

Modality

multimodal

Evaluates

多模态, Multimodal, 推理, Reasoning, 多模态模型, VLM, 图像理解, Image Understanding, 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/HallusionBench

HallusionBench paper and code

• Paper https://arxiv.org/abs/2310.14566
• Paper https://arxiv.org/abs/2406.10900
• Code https://github.com/tianyi-lab/HallusionBench

Which sources cite HallusionBench?

• HallusionBench

Related 多模态, Multimodal, 推理, Reasoning, 多模态模型, VLM, 图像理解, Image Understanding, 逻辑推理, 不支持, Unsupported benchmarks

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