SNAPE-PM: Building and Utilizing Dynamic Partner Models for Adaptive Explanation Generation

TL;DR

This paper introduces SNAPE-PM, a Bayesian and MDP-based framework for dynamically modeling and adapting explanation strategies in dialogue systems to improve personalization and effectiveness.

Contribution

It presents a novel computational partner model and decision process for adaptive explanation generation, addressing the challenge of user-specific and context-aware explanations.

Findings

Effective adaptation to different simulated interlocutors

High accuracy in tracking user feedback and preferences

Emergence of distinct explanation strategies for different partners

Abstract

Adapting to the addressee is crucial for successful explanations, yet poses significant challenges for dialogsystems. We adopt the approach of treating explanation generation as a non-stationary decision process, where the optimal strategy varies according to changing beliefs about the explainee and the interaction context. In this paper we address the questions of (1) how to track the interaction context and the relevant listener features in a formally defined computational partner model, and (2) how to utilize this model in the dynamically adjusted, rational decision process that determines the currently best explanation strategy. We propose a Bayesian inference-based approach to continuously update the partner model based on user feedback, and a non-stationary Markov Decision Process to adjust decision-making based on the partner model values. We evaluate an implementation of this framework with five simulated interlocutors, demonstrating its effectiveness in adapting to different partners with constant and even changing feedback behavior. The results show high adaptivity with distinct explanation strategies emerging for different partners, highlighting the potential of our approach to improve explainable AI systems and dialogsystems in general.