Alfie: An Interactive Robot with a Moral Compass

TL;DR

Alfie is an interactive social robot capable of answering moral questions using advanced language models, and it learns from user interactions to improve its responses.

Contribution

This paper introduces Alfie, a novel social robot that answers moral questions and learns from user feedback using state-of-the-art sentence embedding models.

Findings

Alfie successfully interacts with users on moral topics.

Alfie can adapt its answers based on user disagreement.

The system demonstrates effective use of language models in social robots.

Abstract

This work introduces Alfie, an interactive robot that is capable of answering moral (deontological) questions of a user. The interaction of Alfie is designed in a way in which the user can offer an alternative answer when the user disagrees with the given answer so that Alfie can learn from its interactions. Alfie's answers are based on a sentence embedding model that uses state-of-the-art language models, e.g. Universal Sentence Encoder and BERT. Alfie is implemented on a Furhat Robot, which provides a customizable user interface to design a social robot.