Moral Stories: Situated Reasoning about Norms, Intents, Actions, and their Consequences

TL;DR

This paper introduces 'Moral Stories', a dataset and methods for training AI models to reason about social norms, intentions, and consequences, enabling more socially aware artificial systems.

Contribution

It presents a new dataset and decoding strategies that enhance AI's ability to generate norm-compliant actions, predict outcomes, and explain moral reasoning.

Findings

Decoding strategies improve the quality of generated social reasoning.

Models can generate goal-oriented actions under moral constraints.

The dataset enables grounded, structured social reasoning research.

Abstract

In social settings, much of human behavior is governed by unspoken rules of conduct. For artificial systems to be fully integrated into social environments, adherence to such norms is a central prerequisite. We investigate whether contemporary NLG models can function as behavioral priors for systems deployed in social settings by generating action hypotheses that achieve predefined goals under moral constraints. Moreover, we examine if models can anticipate likely consequences of (im)moral actions, or explain why certain actions are preferable by generating relevant norms. For this purpose, we introduce 'Moral Stories', a crowd-sourced dataset of structured, branching narratives for the study of grounded, goal-oriented social reasoning. Finally, we propose decoding strategies that effectively combine multiple expert models to significantly improve the quality of generated actions, consequences, and norms compared to strong baselines, e.g. though abductive reasoning.