Application of Large Language Model Methods in Scientific and Technical Intelligence Practice

Authors

• Bolin HUA, Department of Information Management, Peking University, Beijing 100871
• Yingze WANG, Department of Information Management, Peking University, Beijing 100871

Keywords

large language model; scientific and technical intelligence; intelligence method; intelligence practice; deeplearning; textual information

Abstract

[Purpose/significance]With the strong ability to process large-scale datasets and outstanding performance in various natural language processing tasks, large language models (LLMs) have excelled across multiple industries.Since scientific and technical intelligence primarily relies on textual data, LLMs are naturally well-suited for this field, ushering in a new wave of transformative changes. [Method /process]This article discusses the advantages of LLMs from five perspectives: low-dimensional dense vector representations of text, large-scale pre-trained models,fine-tuning and prompt learning, high-quality large-scale training data, and human alignment techniques. [Result/conclusion]LLMs have extensive applications in tasks such as intelligence identification, intelligence tracking, intelligence evaluation, and intelligence prediction, resulting in significant optimization improvements or paradigm shifts.

First Page

53

Last Page

64

Submission Date

22-Aug-2024

Revision Date

08-Oct-2024

Acceptance Date

29-Nov-2024

Published Date

01-Jan-2025

Reference

[1] ZHAO W X, ZHOU K, LI J, et al. A Survey of Large Language Models[J]. arXiv preprint arXiv: 2303. 18223, 2023.
[2] 章成志, 苏新宁. 基于条件随机场的自动标引模型研究[J]. 中国图书馆学报, 2008 (05): 89-94, 99.
[3] 杨丽, 吴雨茜, 王俊丽, 等. 循环神经网络研究综述[J]. 计算机应用, 2018, 38 (S2): 1-6, 26.
[4] VASWANI A, SHAZEER N, PARMAR N, et al. Attention Is All You Need[C]//31st Conference on Neural Information Processing Systems, Long Beach, CA: 2017.
[5] DEVLIN J, CHANG M W, LEE K, et al. BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding[C]//Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. Stroudsburg: Association for Computational Linguistics, 2019: 4171-4186.
[6] LIU Y, OTT M, GOYAL N, et al. RoBERTa: A Robustly Optimized BERT Pretraining Approach[J]. arXiv preprint arXiv: 1907. 11692, 2019.
[7] LAN Z, CHEN M, GOODMAN S, et al. AlBert: A Lite Bert For Self-Supervised Learning Of Language Representations[J]. arXiv preprint arXiv: 1909. 11942, 2019.
[8] SANH V, DEBUT L, CHAUMOD J, et al. DistilBERT a distilled version of BERT: smaller, faster, cheaper and lighter[J]. arXiv preprint arXiv: 1910. 01108, 2019.
[9] RADFORD A, NARASIMHAN K, SALIMANS T, et al. Improving Language UnderstandingBy Generative Pre-Training[J]. Computer Science, 2018: 1-12.
[10] 何哲, 曾润喜, 秦维, 等. ChatGPT等新一代人工智能技术的社会影响及其治理[J]. 电子政务, 2023 (04): 2-24.
[11] 李西雨, 钱力, 张智雄. 基于提示微调的科技论文语义评价指标量化方法研究[J]. 数据分析与知识发现, 2024, 8 (Z1): 200-212.
[12] JI J, QIU T, CHEN B, et al. Ai alignment: A comprehensive survey[J]. arXiv preprint arXiv: 2310. 19852, 2023.
[13] WANG X, DUAN S, YI X, et al. On the Essence and Prospect: An Investigation of Alignment Approaches for Big Models[J]. arXiv preprint arXiv: 2403. 04204, 2024.
[14] CHRISTIANO P F, LEIKE J, BROWN T, et al. Deep Reinforcement Learning from Human Preferences[J]. Advances in Neural Information Processing Systems, 2017, 30.
[15] 克拉克. 情报分析: 以目标为中心的方法[M]. 马忠元, 译. 北京: 金城出版社, 2013.
[16] 王延飞, 赵柯然, 陈美华, 等. 情报感知的研究解析[J]. 情报理论与实践, 2018, 41 (08): 1-4.
[17] 化柏林, 李广建. 面向情报流程的情报方法体系构建[J]. 情报学报, 2016, 35 (02): 177-188.
[18] 张智雄, 于改红, 刘熠, 等. ChatGPT对文献情报工作的影响[J]. 数据分析与知识发现, 2023, 7 (03): 36-42.
[19] 赵浜, 曹树金. 国内外生成式AI大模型执行情报领域典型任务的测试分析[J]. 情报资料工作, 2023, 44 (05): 6-17.
[20] 汤珊红, 李晓松, 赵柯然, 等. 生成式人工智能赋能国防科技情报[J]. 情报理论与实践, 2023, 46 (11): 81-85, 99.
[21] 韩旭, 孙亚伟, 赵璐. 体系化人工智能与大语言模型在智能情报场景中的应用[J]. 北京邮电大学学报, 2024, 47 (04): 11-19, 28.
[22] 新华网. 深势科技发布多模态科学大模型Uni-Finder提升科学文献阅读分析效率[EB/OL]. (2023-11-15) [2024-07-29]. https://www.xinhuanet.com/tech/20231115/d5800f26aa724c479f1e1868c4450133/c.html.
[23] 国家科技图书文献中心. 科技文献信息资源服务体系[EB/OL]. [2024-07-29]. https://www.nstl.gov.cn/.
[24] 湖南省科学技术厅. 湖南科技云[EB/OL]. [2024-07-29]. http://hnkjy.kjt.hunan.gov.cn/.
[25] 陆伟, 汪磊, 程齐凯, 等. 数智赋能信息资源管理新路径: 指令工程的概念、内涵和发展[J]. 图书情报知识, 2024, 41 (01): 6-11.
[26] 赵浜, 曹树金. 国内外生成式AI大模型执行情报领域典型任务的测试分析[J]. 情报资料工作, 2023, 44 (05): 6-17.
[27] 张金柱, 王秋月, 仇蒙蒙. 颠覆性技术识别研究进展综述[J]. 数据分析与知识发现, 2022, 6 (07): 12-31.
[28] 赵建, 梁爽. 关键核心技术识别方法研究进展[J]. 情报杂志, 2024, 43 (04): 68-77.
[29] 张凯, 吕璐成, 韩涛, 等. “论文—专利”关联视角下的新兴技术识别研究[J]. 情报理论与实践, 2024, 47 (09): 183-191.
[30] CHAU M P, ALEXANDER H, SUN S M, et al. TopicGPT: A Prompt-based Topic Modeling Framework. arXiv preprint arXiv: 2311. 01449, 2023.
[31] 王震宇, 朱学芳, 杨睿. 基于多模态大语言模型的关系抽取研究[J/OL]. 数据分析与知识发现, 1-13[2024-11-28]. http://kns.cnki.net/kcms/detail/10.1478.G2.20240117.1110.024.html.
[32] XU D, CHEN W, PENG W, et al. Large Language Models for Generative Information Extraction: A Survey[J]. arXiv preprint arXiv: 2312. 17617, 2023.
[33] 鲍彤, 章成志. ChatGPT中文信息抽取能力测评: 以三种典型的抽取任务为例[J]. 数据分析与知识发现, 2023, 7 (09): 1-11.
[34] 张颖怡, 章成志, 周毅, 等. 基于ChatGPT的多视角学术论文实体识别: 性能测评与可用性研究[J]. 数据分析与知识发现, 2023, 7 (09): 12-24.
[35] 章成志, 谢雨欣, 张恒. 学术文献全文内容中的方法实体细粒度抽取及演化分析研究[J]. 情报学报, 2023, 42 (08): 952-966.
[36] 石栖, 陈文杰, 胡正银, 等. 面向知识发现的科学实验知识图谱构建研究[J/OL]. 数据分析与知识发现, 1-18[2024-11-28]. http://kns.cnki.net/kcms/detail/10.1478.G2.20240626.0951.002.html.
[37] 陆伟, 罗卓然, 李信. 科技创新评价研究进展[J]. 情报学进展, 2022, 14 (00): 158-187.
[38] 王雅琪, 曹树金. ChatGPT用于论文创新性评价的效果及可行性分析[J]. 情报资料工作, 2023, 44 (05): 28-38.
[39] MEYER J G, URBANOWICZ R J, MARTIN P C N, et al. ChatGPT and Large Language Models in Academia: Opportunities and Challenges[J]. BioData Mining, 2023, 16 (01): 20.
[40] JIN M, WANG S, MA L, et al. Time-llm: Time Series Forecasting by Reprogramming Large Language Models[J]. arXiv preprint arXiv: 2310. 01728, 2023.
[41] 孔德婧, 董放, 陈子婧, 等. 离群专利视角下的新兴技术预测: 基于BERT模型和深度神经网络[J]. 图书情报工作, 2021, 65 (17): 131-141.
[42] 张家惠, 丁敬达. 基于BERTopic和LSTM模型的新兴主题预测研究[J/OL]. 情报科学, 1-18[2024-07-14]. http://kns.cnki.net/kcms/detail/22.1264.G2.20240603.1548.010.html.
[43] WEI J, WANG X, SCHUURMANS D, et al. Chain-of-Thought Prompting Elicits Reasoning in Large Language Models[J]. Advances In Neural Information Processing Systems, 2022, 35: 24824-24837.
[44] ZHANG Z, ZHANG A, LI M, et al. Automatic Chain of Thought Prompting in Large Language Models[J]. arXiv preprint arXiv: 2210. 03493, 2022.
[45] ZHU Y, WANG X, CHEN J, et al. Llms for Knowledge Graph Construction and Reasoning: Recent Capabilities and Future Opportunities[J]. arXiv preprint arXiv: 2305. 13168, 2023.

Digital Object Identifier (DOI)

10.19809/j.cnki.kjqbyj.2025.01.005

Recommended Citation

HUA, Bolin and WANG, Yingze (2025) "Application of Large Language Model Methods in Scientific and Technical Intelligence Practice," Journal of Scientific Information Research: Vol. 7: Iss. 1, Article 5.
DOI: 10.19809/j.cnki.kjqbyj.2025.01.005
Available at: https://eng.kjqbyj.com/journal/vol7/iss1/5