Synthetic Dialogue Dataset
收藏资源简介:
Synthetic Dialogue Dataset是由麦考瑞大学创建的数据集,包含476条对话,用于训练和评估目标导向的对话代理。该数据集通过提示工程构建,模拟用户与对话代理之间的交流,旨在从用户处提取信息以生成线性规划模型。数据集内容涵盖多种线性问题描述,来源于NL4Opt数据集,适用于资源分配、计划、调度等多个应用领域,旨在帮助非专业用户解决线性模型构建难题。
The Synthetic Dialogue Dataset, developed by Macquarie University, is a corpus consisting of 476 dialogues intended for training and evaluating goal-oriented dialogue agents. Constructed through prompt engineering, this dataset simulates interactions between users and dialogue agents, with the core objective of extracting necessary information from users to generate linear programming models. Covering a variety of linear problem descriptions sourced from the NL4Opt dataset, the dataset applies to multiple application fields including resource allocation, planning and scheduling, and aims to assist non-expert users in addressing the challenges associated with linear model construction.

- 1Synthetic Dialogue Dataset Generation using LLM Agents麦考瑞大学 · 2024年



