Pragmatic Inference for Moral Reasoning Acquisition: Generalization via Distributional Semantics
This work addresses the problem of improving moral reasoning generalization in LLMs for AI ethics applications, but it appears incremental as it builds on existing moral foundations theory.
The paper tackled the challenge of achieving generalization in moral reasoning for Large Language Models (LLMs) by proposing pragmatic inference methods based on moral foundations theory, resulting in significant enhancement of LLMs' generalization in moral reasoning.
Moral reasoning has emerged as a promising research direction for Large Language Models (LLMs), yet achieving generalization remains a central challenge. From a linguistic standpoint, this difficulty arises because LLMs are adept at capturing distributional semantics, which fundamentally differs from the morals which operate at the pragmatic level. This paper investigates how LLMs can achieve generalized moral reasoning despite their reliance on distributional semantics. We propose pragmatic inference methods grounded in moral foundations theory, which leverage contextual information at each step to bridge the pragmatic gap and guide LLMs in connecting moral foundations with moral reasoning objectives. Experimental results demonstrate that our approach significantly enhances LLMs' generalization in moral reasoning, providing a foundation for future research grounded in moral foundations theory.