Saved in:
Bibliographic Details
Main Authors: Bai, Zilong, Zhang, Ruiji, Chen, Linqing, Cai, Qijun, Zhong, Yuan, Wang, Cong, Fang, Yan, Fang, Jie, Sun, Jing, Wang, Weikuan, Zhou, Lizhi, Hua, Haoran, Qiu, Tian, Wang, Chaochao, Sun, Cheng, Lu, Jianping, Wang, Yixin, Xia, Yubin, Hu, Meng, Liu, Haowen, Xu, Peng, Xu, Licong, Bian, Fu, Gu, Xiaolong, Zhang, Lisha, Wang, Weilei, Tu, Changyang
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2404.18255
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916274236293120
author Bai, Zilong
Zhang, Ruiji
Chen, Linqing
Cai, Qijun
Zhong, Yuan
Wang, Cong
Fang, Yan
Fang, Jie
Sun, Jing
Wang, Weikuan
Zhou, Lizhi
Hua, Haoran
Qiu, Tian
Wang, Chaochao
Sun, Cheng
Lu, Jianping
Wang, Yixin
Xia, Yubin
Hu, Meng
Liu, Haowen
Xu, Peng
Xu, Licong
Bian, Fu
Gu, Xiaolong
Zhang, Lisha
Wang, Weilei
Tu, Changyang
author_facet Bai, Zilong
Zhang, Ruiji
Chen, Linqing
Cai, Qijun
Zhong, Yuan
Wang, Cong
Fang, Yan
Fang, Jie
Sun, Jing
Wang, Weikuan
Zhou, Lizhi
Hua, Haoran
Qiu, Tian
Wang, Chaochao
Sun, Cheng
Lu, Jianping
Wang, Yixin
Xia, Yubin
Hu, Meng
Liu, Haowen
Xu, Peng
Xu, Licong
Bian, Fu
Gu, Xiaolong
Zhang, Lisha
Wang, Weilei
Tu, Changyang
contents In recent years, large language models(LLMs) have attracted significant attention due to their exceptional performance across a multitude of natural language process tasks, and have been widely applied in various fields. However, the application of large language models in the Intellectual Property (IP) domain is challenging due to the strong need for specialized knowledge, privacy protection, processing of extremely long text in this field. In this technical report, we present for the first time a low-cost, standardized procedure for training IP-oriented LLMs, meeting the unique requirements of the IP domain. Using this standard process, we have trained the PatentGPT series models based on open-source pretrained models. By evaluating them on the open-source IP-oriented benchmark MOZIP, our domain-specific LLMs outperforms GPT-4, indicating the effectiveness of the proposed training procedure and the expertise of the PatentGPT models in the IP domain. Remarkably, our model surpassed GPT-4 on the 2019 China Patent Agent Qualification Examination, scoring 65 and matching human expert levels. Additionally, the PatentGPT model, which utilizes the SMoE architecture, achieves performance comparable to that of GPT-4 in the IP domain and demonstrates a better cost-performance ratio on long-text tasks, potentially serving as an alternative to GPT-4 within the IP domain.
format Preprint
id arxiv_https___arxiv_org_abs_2404_18255
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle PatentGPT: A Large Language Model for Intellectual Property
Bai, Zilong
Zhang, Ruiji
Chen, Linqing
Cai, Qijun
Zhong, Yuan
Wang, Cong
Fang, Yan
Fang, Jie
Sun, Jing
Wang, Weikuan
Zhou, Lizhi
Hua, Haoran
Qiu, Tian
Wang, Chaochao
Sun, Cheng
Lu, Jianping
Wang, Yixin
Xia, Yubin
Hu, Meng
Liu, Haowen
Xu, Peng
Xu, Licong
Bian, Fu
Gu, Xiaolong
Zhang, Lisha
Wang, Weilei
Tu, Changyang
Computation and Language
Artificial Intelligence
I.2.7
In recent years, large language models(LLMs) have attracted significant attention due to their exceptional performance across a multitude of natural language process tasks, and have been widely applied in various fields. However, the application of large language models in the Intellectual Property (IP) domain is challenging due to the strong need for specialized knowledge, privacy protection, processing of extremely long text in this field. In this technical report, we present for the first time a low-cost, standardized procedure for training IP-oriented LLMs, meeting the unique requirements of the IP domain. Using this standard process, we have trained the PatentGPT series models based on open-source pretrained models. By evaluating them on the open-source IP-oriented benchmark MOZIP, our domain-specific LLMs outperforms GPT-4, indicating the effectiveness of the proposed training procedure and the expertise of the PatentGPT models in the IP domain. Remarkably, our model surpassed GPT-4 on the 2019 China Patent Agent Qualification Examination, scoring 65 and matching human expert levels. Additionally, the PatentGPT model, which utilizes the SMoE architecture, achieves performance comparable to that of GPT-4 in the IP domain and demonstrates a better cost-performance ratio on long-text tasks, potentially serving as an alternative to GPT-4 within the IP domain.
title PatentGPT: A Large Language Model for Intellectual Property
topic Computation and Language
Artificial Intelligence
I.2.7
url https://arxiv.org/abs/2404.18255