Dataset Card for Moral Dilemma Simulation (MDS)
1. Dataset Description
The Moral Dilemma Simulation (MDS) is a multi-modal benchmark grounded in Moral Foundation Theory (MFT). It is designed to evaluate the moral decision-making processes of Vision-Language Models (VLMs) when facing moral dilemmas. The dataset specifically diagnoses the visual distraction phenomenon—how visual inputs can bypass text-based safety mechanisms and alter a model's moral reasoning.… See the full description on the dataset page: https://huggingface.co/datasets/sodacheer/MDS.
