VizXP: A Visualization Framework for Conveying Explanations to Users in Model Reconciliation Problems

Abstract

Advancements in explanation generation for automated planning algorithms have moved us a step closer towards realizing the full potential of human-AI collaboration in real-world planning applications. Within this context, a framework called model reconciliation has gained a lot of traction, mostly due to its deep connection with a popular theory in human psychology, known as the theory of mind. Existing literature in this setting, however, has mostly been constrained to algorithmic contributions for generating explanations. To the best of our knowledge, there has been very little work on how to effectively convey such explanations to human users, a critical component in human-AI collaboration systems. In this paper, we set out to explore to what extent visualizations are an effective candidate for conveying explanations in a way that can be easily understood. Particularly, by drawing inspiration from work done in visualization systems for classical planning, we propose a visualization framework for visualizing explanations generated from model reconciliation algorithms. We demonstrate the efficacy of our proposed system in a comprehensive user study, where we compare our framework against a text-based baseline for two types of explanations – domain-based and problem-based explanations. Results from the user study show that users, on average, understood explanations better when they are conveyed via our visualization system compared to when they are conveyed via a text-based baseline.

PLEASE: Generating Personalized Explanations in Human-Aware Planning

“…It has been shown that explanations as model reconciliation, presented mostly as text, serve an important and intuitive way of explaining plans to users [4,29]. Additionally, Kumar et al [15] showed that conveying explanations through visualizations tend to be more preferred by users than text alone. On that premise, and given the logical nature of our framework, we ought to say that we do not aim at presenting explanations to users in a logical format.…”

A Logic-Based Framework for Explainable Agent Scheduling Problems

“…However, the accuracy of such translations will need to be validated through additional research as LLMs have been shown to have hallucination issues [40]. Another approach is through visualization systems [22,33], though these systems will likely need to be crafted with significant domain expertise.…”

Actor-Focused Interactive Visualization for AI Planning

“…Additionally, as plans grow larger in size and complexity, navigating through each action and effect individually may quickly become imprac-tical. Kumar et al (2021) proposed a visual system for connecting actions and representing system states applied to model reconciliation, but it also lacks scalability as plan size and complexity grows larger. Chakraborti et al (2017) presented a visualization-enabled system for XAIP that offered summarization and reasoning, but mainly for understanding relationships between sensorial data gathered by the system and planner outputs.…”