Contextualized moral inference
This addresses the critical issue of developing moral awareness in AI systems, though it is incremental as it builds on existing work in contextualized language models.
The paper tackled the problem of automated inference of everyday moral situations by predicting people's intuitive judgments of moral vignettes using a text-based approach, showing that contextualized representations outperform word embeddings and emotion sentiment with substantial advantages across three datasets.
Developing moral awareness in intelligent systems has shifted from a topic of philosophical inquiry to a critical and practical issue in artificial intelligence over the past decades. However, automated inference of everyday moral situations remains an under-explored problem. We present a text-based approach that predicts people's intuitive judgment of moral vignettes. Our methodology builds on recent work in contextualized language models and textual inference of moral sentiment. We show that a contextualized representation offers a substantial advantage over alternative representations based on word embeddings and emotion sentiment in inferring human moral judgment, evaluated and reflected in three independent datasets from moral psychology. We discuss the promise and limitations of our approach toward automated textual moral reasoning.