在模型弱点分析阶段,本文提出算法采用小样本进行特征学习,对学生模型进行训练。设构建标注训练样本集Clabel_train样本数K个,构建标注测试样本集Clabel_test样本数M个,构建标注样本集Clabel样本数为N,其中N = M + K。通过小样本模型训练,完成对学生模型的初始化构建,表示为modelstudentk。基于modelstudentk对标注样本集Clabel进行预测,并对每一条样本数据的错误价值Errvalue进行计算评估,其中Errvalue的计算公式如下:
References
黄震华, 杨顺志, 林威, 等. 知识蒸馏研究综述[J]. 计算机学报, 2022, 45(3): 624-653.
朱炫鹏, 姚海东, 刘隽, 等. 大语言模型算法演进综述[J]. 中兴通讯技术, 2024, 30(2): 9-20.
张钦彤, 王昱超, 王鹤羲, 等. 大语言模型微调技术的研究综述[J]. 计算机工程与应用, 2024, 60(17): 17-33.
Liu, P.F., Yuan, W.Z., Fu, J.L., et al. (2021) Pre-Train Prompt and Predict: A Systematic Survey of Prompting Methods in Natural Language Processing.
Chen, S., Chen, S., Xie, G., Shu, X., You, X. and Li, X. (2024) Rethinking Attribute Localization for Zero-Shot Learning. Science China Information Sciences, 67, 184-196. >https://doi.org/10.1007/s11432-023-4051-9
赵凯琳, 靳小龙, 王元卓. 小样本学习研究综述[J]. 软件学报, 2021, 32(2): 349-369.
周孝青, 段湘煜, 俞鸿飞, 等. 基于递进式半知识蒸馏的神经机器翻译[J]. 中文信息学报, 2021, 35(2): 52-60.
俞亮, 魏永丰, 罗国亮, 等. 基于知识蒸馏的隐式篇章关系识别[J]. 计算机科学, 2021, 48(11): 319-326.
黄友文, 魏国庆, 胡燕芳. DistillBIGRU: 基于知识蒸馏的文本分类模型[J]. 中文信息学报, 2022, 36(4): 81-89.
廖胜兰, 吉建民, 俞畅, 等. 基于BERT模型与知识蒸馏的意图分类方法[J]. 计算机工程, 2021, 47(5): 73-79.
顾佼佼, 翟一琛, 姬嗣愚, 等. 基于BERT和知识蒸馏的航空维修领域命名实体识别[J]. 电子测量技术, 2023, 46(3): 19-24.
Shi, C., Su, J., Chu, C., Wang, B. and Feng, D. (2024) Balancing Privacy and Robustness in Prompt Learning for Large Language Models. Mathematics, 12, Article No. 3359. >https://doi.org/10.3390/math12213359
Feng, K., Luo, L., Xia, Y., Luo, B., He, X., Li, K., et al. (2024) Optimizing Microservice Deployment in Edge Computing with Large Language Models: Integrating Retrieval Augmented Generation and Chain of Thought Techniques. Symmetry, 16, Article No. 1470. >https://doi.org/10.3390/sym16111470
Do, D., Nguyen, M. and Nguyen, L. (2025) Enhancing Zero-Shot Multilingual Semantic Parsing: A Framework Leveraging Large Language Models for Data Augmentation and Advanced Prompting Techniques. Neurocomputing, 618, Article ID: 129108. >https://doi.org/10.1016/j.neucom.2024.129108
Ullah, F., Gelbukh, A., Zamir, M.T., Riverόn, E.M.F. and Sidorov, G. (2024) Enhancement of Named Entity Recognition in Low-Resource Languages with Data Augmentation and BERT Models: A Case Study on Urdu. Computers, 13, Article No. 258. >https://doi.org/10.3390/computers13100258
Feng, S.J.H., Lai, E.M. and Li, W. (2024) Geometry of Textual Data Augmentation: Insights from Large Language Models. Electronics, 13, Article No. 3781. >https://doi.org/10.3390/electronics13183781
温浩, 杨洋. 融合ERNIE与知识增强的临床短文本分类研究[J/OL]. 计算机工程与应用, 2025: 1-10. >http://kns.cnki.net/kcms/detail/11.2127.TP.20240527.1040.004.html, 2025-01-15.
杨笑笑, 陆奎. 融合ERNIE和深度学习的文本分类方法[J]. 湖北民族大学学报(自然科学版), 2023, 41(4): 506-512.
Lu, Y.J., Liu, Q., Dai, D., et al. (2022) Unified Structure Generation for Universal Information Extraction.
Ye, Z., Qi, D., Liu, H., Yan, Y., Chen, Q. and Liu, X. (2024) Rouie: A Method for Constructing Knowledge Graph of Power Equipment Based on Improved Universal Information Extraction. Energies, 17, Article No. 2249. >https://doi.org/10.3390/en17102249
Zhang, J., Zhao, Y., Saleh, M., et al. (2020) Pegasus: Pre-Training with Extracted Gap-Sentences for Abstractive Summarization. International Conference on Machine Learning. PMLR, 13-18 July 2020, 11328-11339.
Wang, J., Zhang, Y., Zhang, L., et al. (2022) Fengshenbang 1.0: Being the Foundation of Chinese Cognitive Intelligence.