arXiv cs.AI· Van Bach Nguyen, Qianli Wang, Nils Feldhus, J\"org Schl\"otterer, Christin Seifert·· 4 小时前AI 评分35
SimXAI：一个简单即插即用层提升对话式 XAI 的交互质量与可理解性
SimXAI: A Simple Drop-in Layer Increases Interaction Quality and Understanding in Conversational XAI
AI 导读
SimXAI 是一个轻量的 LLM 对话层，用单个支持工具调用的 LLM 取代 GUS 式架构中意图识别、对话管理和模板生成等模块，以模块化接口接入既有 XAI 资源。它配套推出覆盖单轮/多轮解析、忠实度和上下文化的评测套件，在现有及新数据集上取得较强的功能选择准确率和高忠实度。在 TalkToModel 的用户研究中，仅替换对话层后用户完成任务更准确且轮次更少。
正文 · 原文
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.
| Comments: | Accepted at EMNLP 2026 Findings |
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC) |
| Cite as: | arXiv:2610.41856 [cs.CL] |
| (or arXiv:2610.41856v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.41856 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Qianli Wang [view email]
[v1]
Fri, 30 Oct 2026 15:11:47 UTC (2,693 KB)
来源：arXiv cs.AI · arxiv.org