国创智能家电高质量数据集

包含60万条家电指令的高质量微调数据集，经过意图识别、槽位标注、句式泛化等精细处理，最终生成结构化标注数据，可高效赋能大模型在智能家电语音控制、指令精准解析、多设备联动交互等场景的应用落地与能力优化

This is a high-quality fine-tuning dataset containing 600,000 smart home appliance instruction samples. It has undergone fine-grained processing including intent recognition, slot tagging, sentence pattern generalization and other relevant steps, ultimately generating structured annotated data. This dataset can effectively empower large language models (LLMs) to achieve practical application deployment and optimize their capabilities in scenarios including smart home appliance voice control, precise instruction parsing, and multi-device linkage interaction.

搜集汇总

背景与挑战

背景概述

该数据集是一个包含60万条家电指令的高质量微调数据集，经过意图识别、槽位标注和句式泛化等精细处理，生成了结构化标注数据。它旨在赋能大模型在智能家电语音控制、指令解析和多设备交互等场景的应用与优化。