Mediators: Conversational Agents Explaining NLP Model Behavior
Computation and Language
2022-06-14 v1 Artificial Intelligence
Human-Computer Interaction
Machine Learning
Abstract
The human-centric explainable artificial intelligence (HCXAI) community has raised the need for framing the explanation process as a conversation between human and machine. In this position paper, we establish desiderata for Mediators, text-based conversational agents which are capable of explaining the behavior of neural models interactively using natural language. From the perspective of natural language processing (NLP) research, we engineer a blueprint of such a Mediator for the task of sentiment analysis and assess how far along current research is on the path towards dialogue-based explanations.
Cite
@article{arxiv.2206.06029,
title = {Mediators: Conversational Agents Explaining NLP Model Behavior},
author = {Nils Feldhus and Ajay Madhavan Ravichandran and Sebastian Möller},
journal= {arXiv preprint arXiv:2206.06029},
year = {2022}
}
Comments
Accepted to IJCAI-ECAI 2022 Workshop on Explainable Artificial Intelligence (XAI)