预训练大语言模型发展对中国数字创意产业的启示  

Enlightenment of the Development of Pre-Training Large Language Model to the Digital Creative Industry of China

作　　者：魏晓[1] 陈茂清[2] 曹小琴[1] 许芳婷 Wei Xiao;Chen Maoqing;Cao Xiaoqin;Xu Fangting(Zhongkai University of Agriculture and Engineering,Guangzhou 510225,China;Guangdong Productivity Center,Guangzhou 510070,China)

机构地区：[1]仲恺农业工程学院,广东广州510225 [2]广东省生产力促进中心,广东广州510070

出　　处：《科技管理研究》2024年第12期123-129,共7页Science and Technology Management Research

基　　金：广东省本科高校教学质量与教学改革工程建设项目“基于‘产教融合、校企合作’的产品设计专业创新创业人才培养模式研究与实践”(KA220160158)。

摘　　要：预训练大语言模型与数字创意产业的结合,一方面可以促进预训练大语言模型技术研发和应用,推动自然语言处理相关产业发展,另一方面也可以为数字创意产业提供更高效、精准的解决方案,促进产业数字化转型升级。然而,目前中国预训练大语言模型在数字创意产业的运用主要侧重于文本识别生成和语音生成等领域。为此,通过阐述预训练大语言模型以及中国数字创意产业的发展现状,梳理预训练大语言模型在数字创意产业的应用范畴和商业布局,综合分析作为新质生产力引擎的预训练大语言模型在中国数字创意产业发展中的机遇与挑战,并为中国数字创意产业的发展提出建议。研究发现:融合发展是中国数字创意产业的重要趋势,网络文学、动漫游戏、短视频等细分产业开始发展出完整的产业链条;预训练大语言模型可提升数字创意产业的内容生成效率、丰富艺术创意、拓展数字娱乐形式,也可以加强社交媒体分析监测、提高跨语言应用的效率、辅助科研教育,带来提升数字创意产业的智能化水平、增强用户黏性、数字创意生产者身份多元化等机遇,但同时也面临数据成本、隐私安全、知识产权等问题。提出未来在预训练大语言模型应用于数字创意产业的发展中,重视构建相关监管评估框架和知识产权保护体系,提升多模态技术水平,强化智能算力体系建设,以推动数字创意产业的智能化发展。The combination of pre-training large language model and digital creative industry,on the one hand,can promote the development and application of pre-training large language model technology and promote the development of natural language processing related industries,on the other hand,it can also provide more efficient and accurate solutions for the digital creative industry and promote the digital transformation and upgrading of the industry.However,at present,the application of pre-trained large language models in the digital creative industry mainly focuses on the fields of text recognition generation and speech generation.To this end,this paper expounds the pre-trained large language model and the development status of China's digital creative industry,and sorts out the application scope and business layout of the model in the digital creative industry.Through comprehensive analysis of the opportunities and challenges of the pre-trained large language model as a new quality productivity engine in China's digital creative industry,some suggestions are put forward.The research finds that:integrative development is an important trend of China's digital creative industry.Subdivided industries like online literature,animation games,short videos,begin to develop a complete industrial chain.Pre-trained large language models can improve the content generation efficiency of digital creative industries,enrich artistic creativity,increase digital entertainment forms,strengthen social media analysis and monitoring,improve the efficiency of cross-language applications,assist scientific research and education,and bring opportunities to improve the intelligence level of digital creative industries,enhance user stickiness,and diversify the identity of digital creative producers.At the same time,it also faces problems such as data cost,privacy and security,and intellectual property rights.It is proposed that in the future,the pre-training large language model should be applied to the development of digital creative industry,

关 键 词：大语言模型 预训练模型 数字创意产业 自然语言处理技术 文本生成 人工智能 产业智能化 融合发展 

分 类 号：F49[经济管理—产业经济] TP18[自动化与计算机技术—控制理论与控制工程] G301[自动化与计算机技术—控制科学与工程]