DeepSeek技术创新与通用人工智能发展趋势

吴文峻; 廖星创; 赵金琨

doi:10.3981/j.issn.1000-7857.2025.02.00175

科技导报 >

2025 , Vol. 43 >Issue 6: 14 - 20

DOI: https://doi.org/10.3981/j.issn.1000-7857.2025.02.00175

科技评论

DeepSeek技术创新与通用人工智能发展趋势

吴文峻 ^,¹^,² ,
廖星创 ¹ ,
赵金琨 ¹

展开

1. 北京航空航天大学复杂关键软件环境全国重点实验室, 北京 100191
2. 杭州市北京航空航天大学国际创新研究院, 杭州 311115

吴文峻, 教授, 研究方向为可信智能、群体智能、AI for Science, 电子信箱: wwj09315@buaa.edu.cn

收稿日期: 2025-02-14

网络出版日期: 2025-04-19

版权

收起

DeepSeek: Technological innovations and development trends toward artificial general intelligence

Wenjun WU ^,¹^,² ,
Xingchuang LIAO ¹ ,
Jinkun ZHAO ¹

Expand

1. State Key Laboratory of Complex and Critical Software Environment, Beihang University, Beijing 100191, China
2. International Innovation Institute, Beihang University, Hangzhou 311115, China

Received date: 2025-02-14

Online published: 2025-04-19

Copyright

Fold

摘要

概述了DeepSeek在通用人工智能领域的最新进展，重点讨论了其在大语言模型、推理技术方面的创新。DeepSeek-V3引入了新的模型架构和算法设计，基于相对有限的智能硬件，对模型训练方法进行了全面和深入的优化，显著提升了模型训练效率。在推理技术方面，DeepSeek-R1创新性地结合了强化学习（RL）与监督微调（SFT），提升了推理深度和逻辑推理能力。结合DeepSeek的创新工作，讨论了通用人工智能发展趋势，重点涉及3个问题：开源开放生态对发展通用人工智能的作用；依赖于模型规模扩展的“Neural Scaling Law”是否还能发挥作用；如何基于DeepSeek这类基座模型，以“通专结合”的方式实现行业大模型的落地等。

关键词： DeepSeek; 强化学习; 混合专家系统; 规模法则; 慢思考

本文引用格式

吴文峻 , 廖星创 , 赵金琨 . DeepSeek技术创新与通用人工智能发展趋势[J]. 科技导报, 2025 , 43(6) : 14 -20 . DOI: 10.3981/j.issn.1000-7857.2025.02.00175

参考文献

原文顺序 | 文献年度倒序 | 文中引用次数倒序

1	Liu A , Feng B , Xue B , et al. DeepSeek-V3 technical report[J]. Computer Science, 2024, DOI

2	DeepSeek-AI , Guo D Y , Yang D J , et al. DeepSeek-R1:Incentivizing reasoning capability in LLMs via reinforcement learning[J]. Computer Science, 2025, DOI

3	Jaech A , Kalai A , Lerer A , et al. Openai o1 system card[J]. Computer Science, 2024, DOI

4	Vaswani A . Attention is all you need[J]. Advances in Neural Information Processing Systems, 2017, DOI

5	Huang L , Yu W J , Ma W T , et al. A survey on hallucination in large language models: Principles, taxonomy, challenges, and open questions[J]. ACM Transactions on Information Systems, 2025, 43 (2): 1- 55.

6	Shao Z H , Wang P Y , Zhu Q H , et al. DeepSeekMath: Pushing the limits of mathematical reasoning in open lang-uage models[J]. Computer Science, 2024, DOI

7	Xin H J , Ren Z Z , Song J X , et al. DeepSeek-prover-V1.5:Harnessing proof assistant feedback for reinforcement learning and Monte-Carlo tree searc[J]. Computer Science, 2024, DOI

8	Achiam J , Adler S , Agarwal S , et al. Gpt-4 technical report[J]. Computer Science, 2023, DOI

9	Team G , Anil R , Borgeaud S , et al. Gemini: A family of highly capable multimodal models[J]. Computer Science, 2023, DOI

10	Schulman J , Wolski F , Dhariwal P , et al. Proximal policy optimization algorithms[J]. Computer Science, 2017, DOI

11	de Moura L, Ullrich S. The lean 4 theorem prover and programming language[C]//Automated Deduction-CADE 28: 28th International Conference on Automated Deduction, Virtual Event. Cham: Springer International Publishing, 2021: 625-635.

12	Kaplan J , McCandlish S , Henighan T , et al. Scaling laws for neural language models[J]. Computer Science, 2020, DOI

13	Touvron H , Lavril T , Izacard G , et al. LLaMA: Open and efficient foundation language models[J]. Computer Science, 2023, DOI

14	Jones N . The AI revolution is running out of data. What can researchers do?[J]. Nature, 2024, 636 (8042): 290- 292. DOI

15	Liu Z H , Wang Y H , Han K , et al. Post-training quantization for vision transformer[J]. Advances in Neural Information Processing Systems, 2021, 34: 28092- 28103.

Options

文章导航

摘要

本文引用格式

参考文献

联系我们

访问统计

模态框（Modal）标题

摘要

本文引用格式

参考文献

联系我们

访问统计