Skip to content

Dataset · JingyuSun

Counterfactual VLM benchmark data

JingyuSun/counterfactual-vlm-benchmark-data

Dataset for benchmarking counterfactual vision-language models (VLMs).

What's inside

Contains data payloads for the Counterfactual VLM Benchmark. Used alongside the GitHub repository for loading and evaluation.

What you can do with it

  • Evaluating vision-language models on counterfactual data
  • Benchmarking multimodal systems
  • Testing VLM robustness to counterfactual changes
Tasks
Visual question answering
Downloads this month
31.7K
Likes
2

How to load it

from datasets import load_dataset

ds = load_dataset("JingyuSun/counterfactual-vlm-benchmark-data")