HaloQuest

This Dataset introduces challenging pictures for assessing MLLM hallucinations. It includes questions and ground-truth answers. Original Paper: (https://arxiv.org/abs/2407.15680)[https://arxiv.org/abs/2407.15680] Code Repo: (https://github.com/google/haloquest/)[https://github.com/google/haloquest/]

Papers using HaloQuest (3)

PostAlign: Multimodal Grounding as a Corrective Lens for MLLMs2025Generate, but Verify: Reducing Hallucination in Vision-Language Models with Retrospective Resampling2025HaloQuest: A Visual Hallucination Dataset for Advancing Multimodal Reasoning2024 · 5 cites