VISCO Benchmark

What is VISCO?

VISCO aims to evaluate the critique and correction capabilities of VLMs, which are two essential building blocks towards VLM self-improvement. VISCO requires…

Released

2024-12-03

Modality

multimodal

Evaluates

多模态, Multimodal, 推理, Reasoning, 知识, Knowledge, self-critique, VLM, VLM reasoning, 多模态模型, 跨模态推理, Cross-modal Reasoning, 不支持, Unsupported

Openness

unknown

Importer

OpenCompass Hub

Reported scores

0

Source provenance

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

VISCO paper, code and dataset

• Paper https://arxiv.org/abs/2412.02172
• Code https://github.com/PlusLabNLP/VISCO
• Dataset https://huggingface.co/datasets/uclanlp/VISCO

Which sources cite VISCO?

• VISCO

Related 多模态, Multimodal, 推理, Reasoning, 知识, Knowledge, self-critique, VLM, VLM reasoning, 多模态模型, 跨模态推理, Cross-modal Reasoning, 不支持, Unsupported benchmarks

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