[1]Liu A, Feng B, Xue B, et al. Deepseekv3 technical report[EBOL]. 2024 [20250723]. http:splab.sdu.edu.cnDeepSeek_V3.pdf[2]OpenAI. GPT4 system architecture and safety evaluation[J]. Journal of Artificial Intelligence Research, 2023, 68(8): 145[3]Liang P, Wu C, Chen Y. Holistic evaluation of language models[J]. IEEE Trans on Pattern Analysis and Machine Intelligence, 2023, 45(8): 1023410251[4]Ramesh A, Zhang L, Smith J. SAFETYBench: A comprehensive benchmark for evaluating LLM safety[J]. Nature Machine Intelligence, 2024, 6(2): 156170[5]Zhang Y, Wang Q, Li X. TrustGPT: A comprehensive trustworthiness evaluation framework for large language models[J]. Computational Linguistics, 2023, 49(3): 567589[6]Zhou C, Liu M, Chen Z. SafetyEval: A domainspecific safety evaluation framework for Chinese large language models[J]. ACM Trans on Asian and LowResource Language Information Processing, 2023, 22(4): 125[7]苏艳芳, 袁静, 薛俊民. 大模型安全评估体系框架研究[J]. 信息安全研究, 2024, 10(增刊2): 105109[8]韦韬, 刘焱, 翁海琴, 等. 大模型应用可信框架研究[J]. 信息安全研究, 2024, 10(12): 11531159[9]Ferraiolo D F, Sandhu R, Gavrila S, et al. Proposed NIST standard for rolebased access control[J]. ACM Trans on Information and System Security (TISSEC), 2001, 4(3): 224274[10]Wymberry C, Jahankhani H. An approach to measure the effectiveness of the mitre atlas framework in safeguarding machine learning systems against data poisoning attack[C] Cybersecurity and Artificial Intelligence: Transformational Strategies and Disruptive Innovation. Cham: Springer Nature Switzerland, 2024: 81116[11]Symeonidis G, Nerantzis E, Kazakis A, et al. Mlopsdefinitions, tools and challenges[C] Proc of the 12th IEEE Annual Computing and Communication Workshop and Conference (CCWC). Piscataway, NJ: IEEE, 2022: 453460 |