SimXAI: A Simple Drop-in Layer Increases Interaction Quality and Understanding in Conversational XAI

arXiv:2610.41856 · cs.CL, cs.AI, cs.HC · Submitted 2026-10-30 · Read on arXiv

cs.CL, cs.AI, cs.HC

Submitted: 2026-10-30

Updated: 2026-10-30

Comments: Accepted at EMNLP 2026 Findings

Code: https://github.com/bach-mr/SimXAI

License: http://creativecommons.org/licenses/by/4.0/

The gist: Conversational explainable AI (ConvXAI) agents often use GUS-style architectures with separate modules for intent recognition, dialogue management, and template-based response generation.

Terminology

Abstract

Conversational explainable AI (ConvXAI) agents often use GUS-style architectures with separate modules for intent recognition, dialogue management, and template-based response generation. While faithful to the underlying XAI methods, these pipelines are hard to adapt and often struggle with context-dependent follow-up questions. We present SimXAI, a lightweight LLM-centric conversational layer that replaces these modules with a single tool-calling LLM that bridges access to existing XAI resources. Rather than a standalone agent, SimXAI serves as a modular interface that enables conversational access to external XAI resources. We further introduce an evaluation suite for single- and multi-turn parsing, faithfulness, and contextualization. Across existing and new datasets, SimXAI achieves strong function-selection accuracy and high faithfulness while better handling context-dependent requests. In a user study replacing only the conversational layer of TalkToModel, a representative GUS-based agent, users completed tasks more accurately and in fewer turn. Overall, SimXAI improves GUS-based design while preserving faithfulness and improving conversational quality.

Sources

Related papers