HallusionBench
收藏arXiv2025-09-30 收录
官方服务：
资源简介：
该数据集为多模态模型中的幻觉检测提供了一个基准，它被用于评估在实际应用场景中的性能表现，其核心任务是幻觉检测。
This dataset serves as a benchmark for hallucination detection in multimodal models. It is employed to evaluate model performance in real-world application scenarios, and its core task is hallucination detection.
