MFTCXplain is the first multilingual benchmark dataset designed to evaluate the moral reasoning of Large Language Models (LLM) through multi-hop hate speech explanations grounded in Moral Foundations Theory (MFT).

This repository contains code files and documentation pertaining to the Moral Decision Dataset (MDD). The MDD describes real-world cases, associated parameters, and case-based moral decisions.

TriEthix is a novel evaluation framework that systematically benchmarks frontier LLMs across three foundational ethical perspectives: virtue, deontology, and consequentialism in 3 steps: (Step-1) Moral Weights; (Step-2) Moral Consistency; and (Step-3) Moral Reasoning. TriEthix reveals robust moral profiles for AI Safety, Governance, and Welfare.

A multi-perspective ethical analysis system that evaluates actions through major philosophical frameworks.

Analysis of persistent social conformity on risk preferences in moral and economic choices