"IRWOZ 2.0 - A Large Language Model-driven Dialogue Dataset for Industrial Robot Conversations"

"IRWOZ has improved industrial human-robot interaction (HRI) dialogue systems through domain-specific annotations. However, its initial version contains substantial noise in dialogue states and utterances, limiting state-tracking accuracy. We introduce IRWOZ 2.0, which addresses these limitations through large language model (LLM) enhanced generation (Mistral\/Claude-3.5) and quality refinements. Our improveddataset expands to 390 dialogues across 4 industrial domains (Assembly, Delivery, Position, Relocation), featuring manual corrections and automated typo removal. Benchmark experiments on dialogue state tracking demonstrate significant improvements, with GPT-2\u2019s BLEU-4 score increasing from 0.1651 to 0.5604 compared to original IRWOZ."IRWOZ通过领域专属标注，实现了工业人机交互（Human-Robot Interaction, HRI）对话系统的优化升级。然而其初始版本在对话状态与话语文本中存在大量噪声，制约了对话状态跟踪的精度。为此我们推出了IRWOZ 2.0，通过大语言模型（Large Language Model, LLM）增强生成（采用Mistral/Claude-3.5模型）与质量优化来解决上述局限。优化后的数据集规模扩展至390段对话，覆盖装配（Assembly）、配送（Delivery）、定位（Position）、搬迁（Relocation）4个工业领域，并涵盖手动校正与自动化错别字修正流程。在对话状态跟踪任务上开展的基准测试结果显示，模型性能得到显著提升：相较于原始IRWOZ，GPT-2的BLEU-4评分从0.1651提升至0.5604。

搜集汇总

背景与挑战

背景概述

IRWOZ 2.0是一个专为工业机器人对话设计的大语言模型（LLM）生成数据集，包含390个对话，覆盖装配、交付、位置和重新定位四个工业领域。该数据集通过LLM增强生成和质量优化，解决了先前版本的噪声问题，显著提升了对话状态跟踪性能，例如在基准测试中GPT-2的BLEU-4分数从0.1651提高到0.5604。作为首个针对制造业的LLM生成任务导向语料库，它模拟车间工人与机器人的交互，支持工业人机交互（HRI）研究。