人和大模型

人和大模型，是面向智慧政务领域打造垂直领域大模型，提供丰富的行业知识储备，通过预训练+微调+强化反馈学习的方式，提高大模型在政务行业领域的专业性、准确性和可靠性，降低大模型生成的不确定性，从而可以更好的服务于政务领域应用。

The Renhe Large Language Model is a vertical-domain large language model developed specifically for the smart government affairs domain. It is equipped with a rich repository of industry-specific knowledge, and adopts pre-training, fine-tuning and reinforcement learning from human feedback (RLHF) to enhance the professionalism, accuracy and reliability of the LLM in the government affairs sector, reduce the uncertainty in its generated outputs, and thereby better support applications in the government affairs domain.

搜集汇总

背景与挑战

背景概述

该数据集是针对智慧政务领域构建的垂直大模型，采用预训练、微调和强化反馈学习相结合的方式，旨在提升模型的专业性和准确性，降低生成内容的不确定性，以更好地服务政务应用场景。