持续关系抽取方法研究综述

doi:10.3778/j.issn.1002-8331.2412-0102

摘要/Abstract

摘要： 关系抽取旨在从文本数据中识别并提取实体之间的关系。随着数据流的动态变化，传统关系抽取模型在处理新出现的关系类型时，往往面临灵活性和有效性的双重挑战。持续关系抽取模型通过实时学习，不仅能够适应新关系类型的引入，还能有效保留已学到的知识，为知识图谱的动态更新与扩展提供了重要支持。系统综述了持续关系抽取领域的研究进展。阐述了持续关系抽取的发展历程、基本概念以及任务定义；从关系原型、对抗增强、对比学习及其他方法四个方面总结了当前的研究方法；介绍了常用的数据集与评价指标，并对主流模型的性能进行了对比评估。最后，分析了现有方法的局限性与挑战，并对未来的研究方向提出了展望。

关键词: 持续关系抽取, 记忆机制, 关系原型, 对抗增强, 对比学习

Abstract: Relation extraction (RE) focuses on identifying and extracting relations between entities from textual data. With the dynamic nature of data streams, traditional relation extraction models face challenges with flexibility and effectiveness when new relation types emerge. Continual relation extraction (CRE) models, through real-time learning, not only adapt to new relation types but also effectively retain previously learned knowledge, providing crucial support for the dynamic updating and expansion of knowledge graphs. This paper provides a systematic review of the research progress in the field of continual relation extraction. Firstly, it elaborates the development history, basic concepts, and task definitions of continual relation extraction. Next, it summarizes the current research methods from four perspectives: relation prototype, adversarial augmentation, contrastive learning, and other approaches. Subsequently, commonly used datasets and evaluation metrics are introduced, followed by a comparative performance evaluation of mainstream models. Finally, the limitations and challenges of existing methods are discussed, and future research directions are proposed.

Key words: continual relation extraction, memory mechanism, relation prototypes, adversarial augmentation, contrastive learning

杭婷婷, 郭亚, 李德胜, 冯钧. 持续关系抽取方法研究综述[J]. 计算机工程与应用, 2025, 61(14): 1-19.

HANG Tingting, GUO Ya, LI Desheng, FENG Jun. Survey on Research of Continual Relation Extraction Methods[J]. Computer Engineering and Applications, 2025, 61(14): 1-19.

参考文献

[1] ZARATIANA U, TOMEH N, HOLAT P, et al. An utoregressive text-to-graph framework for joint entity and relation extraction[C]//Proceedings of the AAAI Conference on Artificial Intelligence. Palo Alto: AAAI Press, 2024: 19477-19487.
[2] CUI S Y, CAO J X, CONG X, et al. Enhancing multimodal entity and relation extraction with variational information bottleneck[J]. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2024, 32: 1274-1285.
[3] JIAN Z R, LIU S Q, YIN H X. A multi-granularity contrastive learning for distantly supervised relation extraction[C]//Proceedings of the 20th International Conference on Advanced Intelligent Computing Technology and Applications. Singapore: Springer Nature Singapore, 2024: 352-364.
[4] WANG H, XIONG W, YU M, et al. Sentence embedding alignment for lifelong relation extraction[C]//Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. Wilmington: Association for Computational Linguistics, 2019: 796-806.
[5] SANDERSON K. AI science search engines are exploding in number: are they any good?[J]. Nature, 2023, 616: 639-640.
[6] CHEN X C, YAO L N, MCAULEY J, et al. Deep reinforcement learning in recommender systems: a survey and new perspectives[J]. Knowledge-Based Systems, 2023, 264: 110335.
[7] GOYAL R, KUMAR P, SINGH V P. Automated question and answer generation from texts using text-to-text transformers[J]. Arabian Journal for Science and Engineering, 2024, 49(3): 3027-3041.
[8] WANG L Y, ZHANG X X, SU H, et al. A comprehensive survey of continual learning: theory, method and application[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024, 46(8): 5362-5383.
[9] FRENCH R M. Catastrophic forgetting in connectionist networks[J]. Trends in Cognitive Sciences, 1999, 3(4): 128-135.
[10] LU Y, BARTOLO M, MOORE A, et al. Fantastically ordered prompts and where to find them: overcoming few-shot prompt order sensitivity[C]//Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics. Wilmington: Association for Computational Linguistics, 2022: 8086-8098.
[11] KONG Y J, LIU L, CHEN H H, et al. Overcoming catastrophic forgetting in continual learning by exploring eigenvalues of hessian matrix[J]. IEEE Transactions on Neural Networks and Learning Systems, 2024, 35(11): 16196-16210.
[12] SPARROW Z M, ERNST B G, QUADY T K, et al. Uniting nonempirical and empirical density functional approximation strategies using constraint-based regularization[J]. The Journal of Physical Chemistry Letters, 2022, 13(30): 6896-6904.
[13] HAN Y Z, HUANG G, SONG S J, et al. Dynamic neural networks: a survey[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 44(11): 7436-7456.
[14] CAMBRIA E, COUGHLIN M F, FLORYAN M A, et al. Linking cell mechanical memory and cancer metastasis[J]. Nature Reviews Cancer, 2024, 24(3): 216-228.
[15] NIU Z Y, ZHONG G Q, YU H. A review on the attention mechanism of deep learning[J]. Neurocomputing, 2021, 452: 48-62.
[16] CHICCO D. Siamese neural networks: an overview[J]. Methods in Molecular Biology, 2021, 2190: 73-94.
[17] SUN R, ZHENG Z, WANG Z. Learning encodings for constructive neural combinatorial optimization needs to regret[C]//Proceedings of the AAAI Conference on Artificial Intelligence. Palo Alto: AAAI Press, 2024: 20803-20811.
[18] GUI J, CHEN T, ZHANG J, et al. A survey on self-supervised learning: algorithms, applications, and future trends[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024, 46(12): 9052-9071.
[19] ZHONG W, GUO L, GAO Q, et al. MemoryBank: enhancing large language models with long-term memory[C]//Proceedings of the AAAI Conference on Artificial Intelligence. Palo Alto: AAAI Press, 2024: 19724-19731.
[20] 鄂海红, 张文静, 肖思琪, 等. 深度学习实体关系抽取研究综述[J]. 软件学报, 2019, 30(6): 1793-1818.
E H H, ZHANG W J, XIAO S Q, et al. Survey of entity relationship extraction based on deep learning[J]. Journal of Software, 2019, 30(6): 1793-1818.
[21] 王传栋, 徐娇, 张永. 实体关系抽取综述[J]. 计算机工程与应用, 2020, 56(12): 25-36.
WANG C D, XU J, ZHANG Y. Survey of entity relation extraction[J]. Computer Engineering and Applications, 2020, 56(12): 25-36.
[22] 李冬梅, 张扬, 李东远, 等. 实体关系抽取方法研究综述[J]. 计算机研究与发展, 2020, 57(7): 1424-1448.
LI D M, ZHANG Y, LI D Y, et al. Review of entity relation extraction methods[J]. Journal of Computer Research and Development, 2020, 57(7): 1424-1448.
[23] 韩亚楠, 刘建伟, 罗雄麟. 连续学习研究进展[J]. 计算机研究与发展, 2022, 59(6): 1213-1239.
HAN Y N, LIU J W, LUO X L. Research progress of continual learning[J]. Journal of Computer Research and Development, 2022, 59(6): 1213-1239.
[24] 周大蔚, 汪福运, 叶翰嘉, 等. 基于深度学习的类别增量学习算法综述[J]. 计算机学报, 2023, 46(8): 1577-1605.
ZHOU D W, WANG F Y, YE H J, et al. Deep learning for class-incremental learning: a survey[J]. Chinese Journal of Computers, 2023, 46(8): 1577-1605.
[25] 李文斌, 熊亚锟, 范祉辰, 等. 持续学习的研究进展与趋势[J]. 计算机研究与发展, 2024, 61(6): 1476-1496.
LI W B, XIONG Y K, FAN Z C, et al. Advances and trends of continual learning[J]. Journal of Computer Research and Development, 2024, 61(6): 1476-1496.
[26] 刘壮, 董子宸, 董宜琳, 等. 图终身学习: 综述[J]. 计算机研究与发展, 2024, 61(8): 2067-2096.
LIU Z, DONG Z C, DONG Y L, et al. Lifelong graph learning: a comprehensive review[J]. Journal of Computer Research and Development, 2024, 61(8): 2067-2096.
[27] CHEN L X, LI J P, LUO J L, et al. A review of continual relation extraction[C]//Proceedings of the 2023 20th International Computer Conference on Wavelet Active Media Technology and Information Processing. Piscataway: IEEE, 2023: 1-6.
[28] WAN Q, WEI L N, ZHAO S, et al. A span-based multi-modal attention network for joint entity-relation extraction[J]. Knowledge-Based Systems, 2023, 262: 110228.
[29] WANG L Y, ZHANG X X, SU H, et al. A comprehensive survey of continual learning: theory, method and application[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024, 46(8): 5362-5383.
[30] LANGE D M, ALJUNDI R, MASANA M, et al. A continual learning survey: defying forgetting in classification tasks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 44(7): 3366-3385.
[31] LI Z, HOIEM D. Learning without forgetting[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 40(12): 2935-2947.
[32] TENG S Y, HU X M, DENG P, et al. Motion planning for autonomous driving: the state of the art and future perspectives[J]. IEEE Transactions on Intelligent Vehicles, 2023, 8(6): 3692-3711.
[33] WU C Y, ZHANG X M, ZHANG Y, et al. MedKLIP: medical knowledge enhanced language-image pre-training for X-ray diagnosis[C]//Proceedings of the 2023 IEEE/CVF International Conference on Computer Vision. Piscataway: IEEE, 2023: 21315-21326.
[34] VULLAM N, VELLELA S S, B V R, et al. Multi-agent personalized recommendation system in E-commerce based on user[C]//Proceedings of the 2023 2nd International Conference on Applied Artificial Intelligence and Computing. Piscataway: IEEE, 2023: 1194-1199.
[35] ZONG Y C, ZUO Q K, NG M K, et al. A new brain network construction paradigm for brain disorder via diffusion-based graph contrastive learning[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024, 46(12): 10389-10403.
[36] HOLLY S, MAULIK B, SAMUEL I. Use of whatsapp as a learning media to increase students’ learning interest[J]. Scientechno: Journal of Science and Technology, 2023, 2(1): 35-48.
[37] JAWAHAR G, SAGOT B, SEDDAH D. What does BERT learn about the structure of language?[C]//Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. Stroudsburg: ACL, 2019: 3651-3657.
[38] YIN F F, ZHAO H P, LU S S, et al. DNA-framework-based multidimensional molecular classifiers for cancer diagnosis[J]. Nature Nanotechnology, 2023, 18(6): 677-686.
[39] HAN X, DAI Y, GAO T Y, et al. Continual relation learning via episodic memory activation and reconsolidation[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. Stroudsburg: ACL, 2020: 6429-6440.
[40] WEN M T, XIA T Y, LIAO B W, et al. Few-shot relation classification using clustering-based prototype modification[J]. Knowledge-Based Systems, 2023, 268: 110477.
[41] LIU C, FU X, WANG Y, et al. Overcoming data limitations: a few-shot specific emitter identification method using self-supervised learning and adversarial augmentation[J]. IEEE Transactions on Information Forensics and Security, 2023, 19: 500-513.
[42] WU J, CHEN J, WU J, et al. Understanding contrastive learning via distributionally robust optimization[C]//Proceedings of the 37th International Conference on Neural Information Processing Systems. Berkeley: Neural Information Processing Systems Foundation, 2023: 23297-23320.
[43] DAI M F, XING S, XU Q, et al. Multiprototype relational network for few-shot ALS point cloud semantic segmentation by transferring knowledge from photogrammetric point clouds[J]. IEEE Transactions on Geoscience and Remote Sensing, 2024, 62: 1-17.
[44] MILTON A L. Drug memory reconsolidation: from molecular mechanisms to the clinical context[J]. Translational Psychiatry, 2023, 13(1): 370.
[45] GREENE N R, NAVEH-BENJAMIN M. Adult age-related changes in the specificity of episodic memory representations: a review and theoretical framework[J]. Psychology and Aging, 2023, 38(2): 67-86.
[46] VERSTYNEN T, KORDING K P. Overfitting to ‘predict’ suicidal ideation[J]. Nature Human Behaviour, 2023, 7(5): 680-681.
[47] CUI L, YANG D Q, YU J X, et al. Refining sample embeddings with relation prototypes to enhance continual relation extraction[C]//Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing. Stroudsburg: ACL, 2021: 232-243.
[48] ZHANG H, LIANG B, YANG M, et al. Prompt-based prototypical framework for continual relation extraction[J]. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2022, 30: 2801-2813.
[49] SIMO-SERRA E, IIZUKA S, ISHIKAWA H. Mastering sketching: adversarial augmentation for structured prediction[J]. ACM Transactions on Graphics (TOG), 2018, 37(1): 1-13.
[50] WANG P, SONG Y, LIU T, et al. Learning robust representations for continual relation extraction via adversarial class augmentation[C]//Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing. Somerville: Association for Computational Linguistics, 2022: 6264-6278.
[51] ZHAO W Z, CUI Y N, HU W. Improving continual relation extraction by distinguishing analogous semantics[C]//Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics. Stroudsburg: ACL, 2023: 1162-1175.
[52] EYSENBACH B, ZHANG T, LEVINE S, et al. Contrastive learning as goal-conditioned reinforcement learning[C]//Advances in Neural Information Processing Systems, 2022: 35603-35620.
[53] HU C, YANG D, JIN H, et al. Improving continual relation extraction through prototypical contrastive learning[C]//Proceedings of the 29th International Conference on Computational Linguistics. Beijing: International Committee on Computational Linguistics, 2022: 1885-1895.
[54] ZHAO K, XU H, YANG J G, et al. Consistent representation learning for continual relation extraction[C]//Proceedings of the Findings of the Association for Computational Linguistics (ACL 2022). Stroudsburg: ACL, 2022: 3402-3411.
[55] HUANG M, XIAO M, WANG L, et al. DP-CRE: continual relation extraction via decoupled contrastive learning and memory structure preservation[C]//Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation. Brussels: European Language Resources Association, 2024: 5338-5349.
[56] WU T T, LI X K, LI Y F, et al. Curriculum-meta learning for order-robust continual relation extraction[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2021: 10363-10369.
[57] XIA H M, WANG P Y, LIU T Y, et al. Enhancing continual relation extraction via classifier decomposition[C]//Proceedings of the Findings of the Association for Computational Linguistics (ACL 2023). Stroudsburg: ACL, 2023: 10053-10062.
[58] HAN X, ZHU H, YU P F, et al. FewRel: a large-scale supervised few-shot relation classification dataset with state-of-the-art evaluation[C]//Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing. Stroudsburg: ACL, 2018: 4803-4809.
[59] ALT C, GABRYSZAK A, HENNIG L. TACRED revisited: a thorough evaluation of the TACRED relation extraction task[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. Stroudsburg: ACL, 2020: 1558-1569.
[60] BORDES A, USUNIER N, CHOPRA S, et al. Large-scale simple question answering with memory networks[J]. arXiv: 1506.02075, 2015.
[61] YU M, YIN W, HASAN K S, et al. Improved neural relation detection for knowledge base question answering[C]// Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics. Wilmington: Association for Computational Linguistics, 2017: 571-581.
[62] BEDDARD A, SHERIDAN C E, BARNES M, et al. Improved accuracy average value models of modular multilevel converters[J]. IEEE Transactions on Power Delivery, 2016, 31(5): 2260-2269.
[63] CAO Q, BALASUBRAMANIAN A, BALASUBRAMANIAN N. Towards accurate and reliable energy measurement of NLP models[C]//Proceedings of SustaiNLP: Workshop on Simple and Efficient Natural Language Processing. Wilmington: Association for Computational Linguistics, 2020: 141-148.
[64] DODI? D, REGODI? D. Tokenization and memory optimization for reducing GPU load in NLP deep learning models[J]. Technical Gazette/Tehni?ki Vjesnik, 2024, 31(6).
[65] LORIMER MOSELEY G, LEAKE H B, BEETSMA A J, et al. Teaching patients about pain: the emergence of pain science education, its learning frameworks and delivery strategies[J]. The Journal of Pain, 2024, 25(5): 104425.
[66] BAI B J, YANG X L, LI Y Z, et al. Deep learning-enabled virtual histological staining of biological samples[J]. Light, Science & Applications, 2023, 12(1): 57.
[67] MEISTER C, PIMENTEL T, WIHER G, et al. Locally typical sampling[J]. Transactions of the Association for Computational Linguistics, 2023, 11: 102-121.
[68] MANIAS D M, NAOUM-SAWAYA J, SHAMI A, et al. Robust traffic grooming and infrastructure placement in OTN-over-DWDM networks[J]. Journal of Optical Communications and Networking, 2023, 15(8): 553-568.
[69] ZHOU X H, HUANG Y, KE K, et al. Large-size shape memory alloy plates subjected to cyclic tension: towards novel self-centring connections in steel frames[J]. Thin-Walled Structures, 2023, 185: 110591.
[70] LI Z X, SHANG X Y, HE R, et al. No fear of classifier biases: neural collapse inspired federated learning with synthetic and fixed classifier[C]//Proceedings of the 2023 IEEE/CVF International Conference on Computer Vision. Piscataway: IEEE, 2023: 5296-5306.
[71] SHAHBAZI N, LIN Y, ASUDEH A, et al. Representation bias in data: a survey on identification and resolution techniquesJ]. ACM Computing Surveys, 2023, 55(13S): 1-39.
[72] ZHENG C F, LYU X Y, GAO L L, et al. Prototype-based embedding network for scene graph generation[C]//Proceedings of the 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2023: 22783-22792.
[73] ALLENSPACH S, HISS J A, SCHNEIDER G. Neural multi-task learning in drug design[J]. Nature Machine Intelligence, 2024, 6(2): 124-137.
[74] KWON W, LI Z H, ZHUANG S Y, et al. Efficient memory management for large language model serving with PagedAttention[C]//Proceedings of the 29th Symposium on Operating Systems Principles. New York: ACM, 2023: 611-626.
[75] KUMARAN D, HASSABIS D, MCCLELLAND J L. What learning systems do intelligent agents need?complementary learning systems theory updated[J]. Trends in Cognitive Sciences, 2016, 20(7): 512-534.
[76] SORRENTI A, BELLITTO G, SALANITRI F P, et al. Wake-sleep consolidated learning[J]. arXiv:2401.08623, 2024.