arXiv 2306.176742023-06-30cs.CL

X-RiSAWOZ：高质量的端到端多语言对话数据集与少样本智能体

X-RiSAWOZ: High-Quality End-to-End Multilingual Dialogue Datasets and Few-shot Agents

Stanford University（斯坦福大学）；Tianjin University（天津大学）；Microsoft（微软）；Université Paris-Saclay（巴黎萨克雷大学）；International Institute of Information Technology, Hyderabad（海得拉巴国际信息技术学院）；Hanyang University（汉阳大学）；Karya Inc.（Karya公司）

Mehrad Moradshahi, Tianhao Shen, Kalika Bali, Monojit Choudhury, Gaël de Chalendar, Anmol Goel, Sungkyun Kim, Prashant Kodali, Ponnurangam Kumaraguru, Nasredine… 展开作者

Mehrad Moradshahi, Tianhao Shen, Kalika Bali, Monojit Choudhury, Gaël de Chalendar, Anmol Goel, Sungkyun Kim, Prashant Kodali, Ponnurangam Kumaraguru, Nasredine Semmar, Sina J. Semnani, Jiwon Seo, Vivek Seshadri, Manish Shrivastava, Michael Sun, Aditya Yadavalli, Chaobin You, Deyi Xiong, Monica S. Lam

备注：: Accepted by ACL 2023 Findings

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

为降低多语言数据集构建成本，本文通过翻译与人工编辑构建了端到端多语言对话基准 X-RiSAWOZ，开发了加速后期编辑的混合实体对齐工具集，并在零样本和少样本设置下验证了其有效性。

摘要：

任务导向型对话研究由于为新语言创建数据集的成本高昂，主要集中在英语和中文等少数流行语言上。为降低成本，我们对自动翻译的数据进行人工编辑。我们通过将中文 RiSAWOZ 翻译成 4 种语言（英语、法语、印地语、韩语）以及一种印英混合语言，创建了一个新的多语言基准 X-RiSAWOZ。X-RiSAWOZ 每种语言包含超过 18,000 条经过人工验证的对话话语，并且与以往大多数多语言工作不同，它是一个用于构建功能完备智能体的端到端数据集。在创建 X-RiSAWOZ 时遇到的诸多困难促使我们开发了一套工具集，以加速翻译后新语言数据集的后期编辑。该工具集通过结合神经与基于字典方法的混合实体对齐技术改进了机器翻译，并提供了许多自动化和半自动化的验证检查。我们通过在零样本和少样本设置下训练对话智能体，为 X-RiSAWOZ 建立了强大的基线，在这些设置中，目标语言中可用的标准数据有限。结果表明，我们的翻译和后期编辑方法及工具集可用于经济高效地创建新的高质量多语言对话智能体。我们的数据集、代码和工具包已开源发布。

英文摘要：

Task-oriented dialogue research has mainly focused on a few popular languages like English and Chinese, due to the high dataset creation cost for a new language. To reduce the cost, we apply manual editing to automatically translated data. We create a new multilingual benchmark, X-RiSAWOZ, by translating the Chinese RiSAWOZ to 4 languages: English, French, Hindi, Korean; and a code-mixed English-Hindi language. X-RiSAWOZ has more than 18,000 human-verified dialogue utterances for each language, and unlike most multilingual prior work, is an end-to-end dataset for building fully-functioning agents. The many difficulties we encountered in creating X-RiSAWOZ led us to develop a toolset to accelerate the post-editing of a new language dataset after translation. This toolset improves machine translation with a hybrid entity alignment technique that combines neural with dictionary-based methods, along with many automated and semi-automated validation checks. We establish strong baselines for X-RiSAWOZ by training dialogue agents in the zero- and few-shot settings where limited gold data is available in the target language. Our results suggest that our translation and post-editing methodology and toolset can be used to create new high-quality multilingual dialogue agents cost-effectively. Our dataset, code, and toolkit are released open-source.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。