Zero-shot relation extraction model based on dual contrastive learning

doi:10.11772/j.issn.1001-9081.2024111587

Journal of Computer Applications ›› 2025, Vol. 45 ›› Issue (11): 3555-3563.DOI: 10.11772/j.issn.1001-9081.2024111587

• Artificial intelligence • Previous Articles

Zero-shot relation extraction model based on dual contrastive learning

Bingjie QIU¹, Chaoqun ZHANG¹^,²(), Weidong TANG¹, Bicheng LIANG¹, Danyang CUI¹, Haisheng LUO¹, Qiming CHEN¹

^1.School of Artificial Intelligence，Guangxi Minzu University，Nanning Guangxi 530006，China
^2.Guangxi Key Laboratory of Hybrid Computing and IC Design Analysis （Guangxi Minzu University），Nanning Guangxi 530006，China

Received:2024-11-11 Revised:2025-02-26 Accepted:2025-02-28 Online:2025-03-04 Published:2025-11-10
Contact: Chaoqun ZHANG
About author:QIU Bingjie， born in 2002， M. S. candidate. Her research interests include natural language processing.
TANG Weidong， born in 1968， Ph. D.， professor. His research interests include formal methods， rough data reasoning.
LIANG Bicheng， born in 2000， M. S. candidate. His research interests include image processing.
CUI Danyang， born in 1999， M. S. candidate. His research interests include natural language processing.
LUO Haisheng， born in 2002， M. S. candidate. His research interests include natural language processing.
CHEN Qiming， born in 1999， M. S. candidate. His research interests include natural language processing.
Supported by:
This work is partially supported by National Natural Science Foundation of China(62062011)

基于双重对比学习的零样本关系抽取模型

邱冰婕¹, 张超群¹^,²(), 汤卫东¹, 梁弼诚¹, 崔丹阳¹, 罗海升¹, 陈启明¹

^1.广西民族大学人工智能学院，南宁 530006
^2.广西混杂计算与集成电路设计分析重点实验室（广西民族大学），南宁 530006

通讯作者: 张超群
作者简介:邱冰婕（2002—），女，湖南邵阳人，硕士研究生，主要研究方向：自然语言处理
汤卫东（1968—），男，山东荣成人，教授，博士，主要研究方向：形式化方法、粗糙数据推理
梁弼诚（2000—），男，辽宁朝阳人，硕士研究生，主要研究方向：图像处理
崔丹阳（1999—），男，河南南阳人，硕士研究生，主要研究方向：自然语言处理
罗海升（2002—），男，广东阳江人，硕士研究生，主要研究方向：自然语言处理
陈启明（1999—），男，浙江温州人，硕士研究生，主要研究方向：自然语言处理。
基金资助:
国家自然科学基金资助项目(62062011)

Abstract

Abstract:

To address the issues of overlapping relation representations and incorrect relation predictions in Zero-Shot Relation Extraction （ZSRE） caused by similar entities or relations， a Dual Contrastive Learning-based Zero-Shot Relation Extraction （DCL-ZSRE） model was proposed. Firstly， both instances and relation descriptions were encoded using pre-trained encoders to obtain their vector representations. Secondly， a dual contrastive learning was designed to enhance the distinguishability of relation representations： Instance-level Contrastive Learning （ICL） was used to learn mutual information between instances， then the representations of instances and relation descriptions were concatenated； and Matching-level Contrastive Learning （MCL） was applied to learn the associations between instances and relation descriptions， thereby resolving the problem of overlapping relation representations. Finally， the learned representations from contrastive learning were utilized in the classification module to predict unseen relations. Experimental results on FewRel and Wiki-ZSL datasets demonstrate that DCL-ZSRE model significantly outperforms eight state-of-the-art models in terms of precision， recall， and F1-score， especially with the large number of unseen relation categories. With 15 unseen relations， DCL-ZSRE achieves improvements of 4.76， 4.63 and 4.69 percentage points in three indicators over EMMA （Efficient Multi-grained Matching Approach） model on FewRel dataset， and also achieves improvements of 1.32， 2.20 and 1.76 percentage points on Wiki-ZSL dataset. These results confirm that DCL-ZSRE model effectively distinguishes overlapping relation representations， establishing an efficient and robust approach for ZSRE.

Key words: relation extraction, feature extraction, zero-shot learning, Contrastive Learning (CL), Pre-trained Language Model (PLM)

摘要：

针对零样本关系抽取（ZSRE）中因相似实体或关系导致的关系表示重叠及关系预测错误问题，提出一种基于双重对比学习的零样本关系抽取（DCL-ZSRE）模型。首先，通过预训练编码器对实例和关系描述进行编码，以得到相应的向量表示；其次，设计双重对比学习提高关系表示的可区分度，即通过实例级对比学习（ICL）学习实例之间的互信息，再将实例和关系描述的表示进行连接，并利用匹配级对比学习（MCL）学习实例与关系描述之间的联系，从而解决关系表示重叠问题；最后，根据对比学习中学习到的表示在分类模块对未见关系进行预测。在FewRel和Wiki-ZSL数据集上的实验结果表明，DCL-ZSRE在精确率、召回率和F1值上均明显优于8个先进的对比模型，尤其在未见关系类别较多时：当未见关系类别数为15时，相较于EMMA （Efficient Multi-grained Matching Approach）模型，DCL-ZSRE模型在FewRel数据集上的3项指标分别显著提高了4.76、4.63、4.69个百分点，在Wiki-ZSL数据集上也实现了1.32、2.20、1.76个百分点的增长。DCL-ZSRE模型能有效区分重叠的关系表示，可作为一种有效且鲁棒性强的零样本关系抽取方法。

关键词: 关系抽取, 特征抽取, 零样本学习, 对比学习, 预训练语言模型

CLC Number:

TP391.1

Bingjie QIU, Chaoqun ZHANG, Weidong TANG, Bicheng LIANG, Danyang CUI, Haisheng LUO, Qiming CHEN. Zero-shot relation extraction model based on dual contrastive learning[J]. Journal of Computer Applications, 2025, 45(11): 3555-3563.

邱冰婕, 张超群, 汤卫东, 梁弼诚, 崔丹阳, 罗海升, 陈启明. 基于双重对比学习的零样本关系抽取模型[J]. 《计算机应用》唯一官方网站, 2025, 45(11): 3555-3563.

Figures/Tables 10

Fig. 1 Architecture of DCL-ZSRE model

Fig. 2 Example of encoding process for input instance

Fig. 3 Structure of attention layer

Tab. 1 Experimental parameter settings

参数	取值
训练集批次batchsize_train	32
验证集批次batchsize_valid	640
测试集批次batchsize_test	640
迭代次数epoch	5
文本最大长度max_length	128
学习率lr	0.000 02
候选关系数k	2
实例级温度参数 $τ I C L$	1
匹配级温度参数 $τ M C L$	0.02
损失权重 $α$	0.1
优化器	AdamW^［41］

Tab. 1 Experimental parameter settings

参数	取值
训练集批次batchsize_train	32
验证集批次batchsize_valid	640
测试集批次batchsize_test	640
迭代次数epoch	5
文本最大长度max_length	128
学习率lr	0.000 02
候选关系数k	2
实例级温度参数 $τ I C L$	1
匹配级温度参数 $τ M C L$	0.02
损失权重 $α$	0.1
优化器	AdamW^［41］

Tab. 2 Performance comparison of experimental models on two datasets

m	模型	FewRel			Wiki-ZSL
m	模型	P/%	R/%	F₁/%	P/%	R/%	F₁/%
5	SUMASK^［42］*	78.27	72.55	75.30	75.64	70.96	73.23
	ZS-BERT^［10］*	76.96	78.86	77.90	71.54	72.39	71.96
	RelationPrompt^［43］*	90.15	88.50	89.30	70.66	83.75	76.63
	RE-Matching^［5］*	92.82	92.34	92.58	78.19	78.41	78.30
	EMMA^［7］*	94.87	94.48	94.67	91.32	90.65	90.98
	IPPR^［19］*	88.03	87.14	87.58	86.56	83.31	84.91
	AlignRE^［44］*	93.30	92.90	93.09	83.11	80.30	81.64
	PromptMatch^［45］†	91.14	90.86	91.00	77.39	75.90	76.63
	DCL-ZSRE	94.32	93.62	93.96	95.29	94.20	94.74
10	SUMASK^［42］*	64.77	60.94	62.80	62.31	61.08	61.69
	ZS-BERT^［10］*	56.92	57.59	57.25	60.51	60.98	60.74
	RelationPrompt^［43］*	80.33	79.62	79.96	68.51	74.76	71.50
	RE-Matching^［5］*	83.21	82.64	82.93	74.39	73.54	73.96
	EMMA^［7］*	87.97	86.48	87.22	86.00	84.55	85.27
	IPPR^［19］*	65.65	63.19	64.39	69.21	70.70	69.95
	AlignRE^［44］*	86.41	85.14	85.75	75.00	73.26	74.10
	PromptMatch^［45］†	83.05	82.55	82.80	71.86	71.14	71.50
	DCL-ZSRE	88.66	87.50	88.08	86.48	87.41	86.94
15	SUMASK^［42］*	44.76	41.13	42.87	43.55	40.27	41.85
	ZS-BERT^［10］*	35.54	38.19	36.82	34.12	34.38	34.25
	RelationPrompt^［43］*	74.33	72.51	73.40	63.69	67.93	65.74
	RE-Matching^［5］*	73.80	73.52	73.66	67.31	67.33	67.32
	EMMA^［7］*	80.47	79.73	80.10	78.51	77.63	78.07
	IPPR^［19］*	49.72	49.44	49.58	43.38	44.95	44.15
	AlignRE^［44］*	77.63	77.00	77.31	69.01	67.52	68.26
	PromptMatch^［45］†	72.83	72.10	72.46	62.13	61.76	61.95
	DCL-ZSRE	85.23	84.36	84.79	79.83	79.83	79.83

Fig. 4 F1 scores of DCL-ZSRE model and comparison models on FewRel and Wiki-ZSL datasets

Tab. 3 Ablation experimental results

模型	FewRel			Wiki-ZSL
模型	P	R	F₁	P	R	F₁
w/o ICL	87.97	87.02	87.49	86.30	86.74	86.51
w/o MCL	31.60	19.71	23.90	28.22	20.78	22.85
w/o IMCL	54.22	54.25	53.22	48.81	50.99	49.88
w/o Class	84.87	85.91	86.97	83.36	84.09	83.72
DCL-ZSRE	88.66	87.50	88.08	86.48	87.41	86.94

Fig. 5 Impact of parameter τMCL on performance of DCL-ZSRE model

Tab. 4 Impact of hyperparameter α on F1-score of DCL-ZSRE model

$α$ 值	F1值/%		$α$ 值	F1值/%
$α$ 值	FewRel	Wiki-ZSL	$α$ 值	FewRel	Wiki-ZSL
0.10	88.078	86.940	0.33	87.982	86.934
0.20	87.990	86.844	0.50	87.984	86.874

Tab. 4 Impact of hyperparameter α on F1-score of DCL-ZSRE model

$α$ 值	F1值/%		$α$ 值	F1值/%
$α$ 值	FewRel	Wiki-ZSL	$α$ 值	FewRel	Wiki-ZSL
0.10	88.078	86.940	0.33	87.982	86.934
0.20	87.990	86.844	0.50	87.984	86.874

Tab. 5 Comparison of required parameters and average running time of five runs of DCL-ZSRE and EMMA on two datasets

模型	FewRel		Wiki-ZSL
模型	时间/h	参数量/10⁷	时间/h	参数量/10⁷
EMMA	1.379	220.739	2.386	220.739
DCL-ZSRE	1.380	224.284	2.404	224.284

References 45

[1]	许亮. 面向中文的零样本关系三元组抽取方法研究［D］. 北京：北京交通大学， 2023： 2-6.
	XU L. Research on zero-shot relation triple extraction method for Chinese［D］. Beijing： Beijing Jiaotong University， 2023： 2-6.
[2]	CAUFIELD J H， HEGDE H， EMONET V， et al. Structured Prompt Interrogation and Recursive Extraction of Semantics （SPIRES）： a method for populating knowledge bases using zero-shot learning［J］. Bioinformatics， 2024， 40（3）： No.btae104.
[3]	许亮，张春，张宁，等.融合多Prompt模板的零样本关系抽取模型［J］.计算机应用，2023，43（12）：3668-3675.
	XU L， ZHANG C， ZHANG N， et al. Zero-shot relation extraction model via multi-template fusion in Prompt［J］. Journal of Computer Applications， 2023， 43（12）： 3668-3675.
[4]	GAUTAM S， POP R. FactGenius： combining zero-shot prompting and fuzzy relation mining to improve fact verification with knowledge graphs［C］// Proceedings of the 7th Fact Extraction and VERification Workshop. Stroudsburg： ACL， 2024： 297-306.
[5]	ZHAO J， ZHAN W， ZHAO X， et al. RE-Matching： a fine-grained semantic matching method for zero-shot relation extraction［C］// Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics （Volume 1： Long Papers）. Stroudsburg： ACL， 2023： 6680-6691.
[6]	潘理虎，刘云，谢斌红，等. 基于语义增强的多特征融合小样本关系抽取［J］. 计算机应用研究， 2022， 39（6）： 1663-1667.
	PAN L H， LIU Y， XIE B H， et al. Multi-feature fusion and few-shot relation extraction based on semantic enhancement［J］. Application Research of Computers， 2022， 39（6）： 1663-1667.
[7]	LI S， BAI G， ZHANG Z， et al. Fusion makes perfection： an efficient multi-grained matching approach for zero-shot relation extraction［C］// Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics： Human Language Technologies （Volume 2： Short Papers）. Stroudsburg： ACL， 2024： 79-85.
[8]	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， Volume 1 （Long and Short Papers）. Stroudsburg： ACL， 2019： 4171-4186.
[9]	HAN X， ZHU H， YU P， 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.
[10]	CHEN C Y， LI C T. ZS-BERT： towards zero-shot relation extraction with attribute representation learning［C］// Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics： Human Language Technologies. Stroudsburg： ACL， 2021： 3470-3479.
[11]	LAN Z， CHEN M， GOODMAN S， et al. ALBERT： a lite BERT for self-supervised learning of language representations［EB/OL］. ［2024-10-15］..
[12]	LAMPERT C H， NICKISCH H， HARMELING S. Attribute-based classification for zero-shot visual object categorization［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2014， 36（3）： 453-465.
[13]	LV F， ZHANG J， YANG G， et al. Learning cross-domain semantic-visual relationships for transductive zero-shot learning［J］. Pattern Recognition， 2023， 141： No.109591.
[14]	PRATT S， COVERT I， LIU R， et al. What does a platypus look like？ generating customized prompts for zero-shot image classification［C］// Proceedings of the 2023 IEEE/CVF International Conference on Computer Vision. Piscataway： IEEE， 2023： 15645-15655.
[15]	ZHANG K， JIMÉNEZ GUTIÉRREZ B， SU Y. Aligning instruction tasks unlocks large language models as zero-shot relation extractors［C］// Findings of the Association for Computational Linguistics： ACL 2023. Stroudsburg： ACL， 2023： 794-812.
[16]	PÀMIES M， LLOP J， MULTARI F， et al. A weakly supervised textual entailment approach to zero-shot text classification［C］// Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics. Stroudsburg： ACL， 2023： 286-296.
[17]	NAJAFI S， FYSHE A. Weakly-supervised questions for zero-shot relation extraction［C］// Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics. Stroudsburg： ACL， 2023： 3075-3087.
[18]	XU L， BU X， TIAN X. Dynamic prompt-driven zero-shot relation extraction［J］. IEEE/ACM Transactions on Audio， Speech， and Language Processing， 2024， 32： 2900-2912.
[19]	DUAN B， LIU X， WANG S， et al. Relational representation learning for zero-shot relation extraction with instance prompting and prototype rectification［C］// Proceedings of the 2023 IEEE International Conference on Acoustics， Speech and Signal Processing. Piscataway： IEEE， 2023： 1-5.
[20]	GONG J， ELDARDIRY H. Zero-shot relation classification from side information［C］// Proceedings of the 30th ACM International Conference on Information and Knowledge Management. New York： ACM， 2021： 576-585.
[21]	ZHANG B， XU Y， LI J， et al. SMDM： tackling zero-shot relation extraction with semantic max-divergence metric learning［J］. Applied Intelligence， 2023， 53（6）： 6569-6584.
[22]	琚生根，黄方怡，孙界平. 融合预训练语言模型的成语完形填空算法［J］. 软件学报， 2022， 33（10）： 3793-3805.
	JU S G， HUANG F Y， SUN J P. Idiom cloze algorithm integrating with pre-trained language model［J］. Journal of Software， 2023， 33（10）： 3793-3805.
[23]	LIU Y， OTT M， GOYAL N， et al. RoBERTa： a robustly optimized BERT pretraining approach［EB/OL］. ［2024-10-15］..
[24]	JOSHI M， CHEN D， LIU Y， et al. SpanBERT： improving pre-training by representing and predicting spans［J］. Transactions of the Association for Computational Linguistics， 2020， 8： 64-77.
[25]	MÜLLER M， SALATHÉ M， KUMMERYOLD P E. COVID-Twitter-BERT： a natural language processing model to analyse COVID-19 content on Twitter［J］. Frontiers in Artificial Intelligence， 2023， 6： No.1023281.
[26]	PEETERS R， BIZER C. Dual-objective fine-tuning of BERT for entity matching［J］. Proceedings of the VLDB Endowment， 2021， 14（10）： 1913-1921.
[27]	BELLO A， NG S C， LEUNG M F. A BERT framework to sentiment analysis of Tweets［J］. Sensors， 2023， 23（1）： No.506.
[28]	LIAO W， LIU Z， DAI H， et al. Mask-guided BERT for few-shot text classification［J］. Neurocomputing， 2024， 610： No.128576.
[29]	HU H， WANG X， ZHANG Y， et al. A comprehensive survey on contrastive learning［J］. Neurocomputing， 2024， 610： No.128645.
[30]	CHEN X， FAN H， GIRSHICK R， et al. Improved baselines with momentum contrastive learning［EB/OL］. ［2024-10-16］..
[31]	朱玄烨，孔兵，陈红梅，等. 困难样本采样联合对比增强的深度图聚类［J］. 计算机应用研究， 2024， 41（6）： 1769-1777.
	ZHU X Y， KONG B， CHEN H M， et al. Deep graph clustering with hard sample sampling joint contrastive augmentation［J］. Application Research of Computers， 2024， 41（6）： 1769-1777.
[32]	CHEN T， KORNBLITH S， NOROUZI M， et al. A simple framework for contrastive learning of visual representations［C］// Proceedings of the 37th International Conference on Machine Learning. New York： JMLR.org， 2020： 1597-1607.
[33]	YANG Z， FEI J， TAN Z， et al. CL&CD： contrastive learning and cluster description for zero-shot relation extraction［J］. Knowledge-Based Systems， 2024， 293： No.111652.
[34]	ZHANG K， WU L， LV G， et al. Description-enhanced label embedding contrastive learning for text classification［J］. IEEE Transactions on Neural Networks and Learning Systems， 2024， 35（10）： 14889-14902.
[35]	赵基藤，李国正，汪鹏，等. 基于监督对比重放的持续关系抽取［J］. 中文信息学报， 2023， 37（11）： 60-67， 80.
	ZHAO J T， LI G Z， WANG P， et al. Continual relation extraction via supervised contrastive replay［J］. Journal of Chinese Information Processing， 2023， 37（11）： 60-67， 80.
[36]	WANG S， ZHANG B， XU Y， et al. RCL： relation contrastive learning for zero-shot relation extraction［C］// Findings of the Association for Computational Linguistics： NAACL 2022. Stroudsburg： ACL， 2022： 2456-2468.
[37]	LUO D， GAN Y， HOU R， et al. Synergistic anchored contrastive pre-training for few-shot relation extraction［C］// Proceedings of the 38th AAAI Conference on Artificial Intelligence. Palo Alto： AAAI Press， 2024： 18742-18750.
[38]	高怡，纪焘，吴苑斌，等. 基于标签增强和对比学习的鲁棒小样本事件检测［J］. 中文信息学报， 2023， 37（4）： 98-108.
	GAO Y， JI T， WU Y B， et al. Robust few shot event detection based on label augmentation and contrastive learning［J］. Journal of Chinese Information Processing， 2023， 37（4）： 98-108.
[39]	LIN Z， FENG M， DOS SANTOS C N， et al. A structured self-attentive sentence embedding［EB/OL］. ［2024-10-15］..
[40]	VAN DEN OORD A， LI Y， VINYALS O. Representation learning with contrastive predictive coding［EB/OL］. ［2024-10-19］..
[41]	LOSHCHILOV I， HUTTER F. Decoupled weight decay regularization［EB/OL］. ［2024-10-15］..
[42]	LI G， WANG P， KE W. Revisiting large language models as zero-shot relation extractors［C］// Findings of the Association for Computational Linguistics： EMNLP 2023. Stroudsburg： ACL， 2023： 6877-6892.
[43]	CHIA Y K， BING L， PORIA S， et al. RelationPrompt： leveraging prompts to generate synthetic data for zero-shot relation triplet extraction［C］// Findings of the Association for Computational Linguistics： ACL 2022. Stroudsburg： ACL， 2022： 45-57.
[44]	LI Z， ZHANG F， CHENG J. AlignRE： an encoding and semantic alignment approach for zero-shot relation extraction［C］// Findings of the Association for Computational Linguistics： ACL 2024. Stroudsburg： ACL， 2024： 2957-2966.
[45]	SAINZ O， LOPEZ DE LACALLE O L， LABAKA G， et al. Label verbalization and entailment for effective zero-and few-shot relation extraction［C］// Proceedings of the 2021 Conference on Empirical Methods in Natural Language. Processing. Stroudsburg： ACL， 2021： 1199-1212.

Zero-shot relation extraction model based on dual contrastive learning

基于双重对比学习的零样本关系抽取模型

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 10

References 45

Related Articles 15

Recommended Articles

Metrics

[1]	Chuang WANG, Lu YU, Jianwei CHEN, Cheng PAN, Wenbo DU. Review of open set domain adaptation [J]. Journal of Computer Applications, 2025, 45(9): 2727-2736.
[2]	Zhiyuan WANG, Tao PENG, Jie YANG. Integrating internal and external data for out-of-distribution detection training and testing [J]. Journal of Computer Applications, 2025, 45(8): 2497-2506.
[3]	Jiaxin YAN, Yanping CHEN, Weizhe YANG, Ruizhang HUANG, Yongbin QIN. Heterogeneous graph attention network for relation extraction based on feature combination [J]. Journal of Computer Applications, 2025, 45(8): 2470-2476.
[4]	Wei ZHANG, Jiaxiang NIU, Jichao MA, Qiongxia SHEN. Chinese spelling correction model ReLM enhanced with deep semantic features [J]. Journal of Computer Applications, 2025, 45(8): 2484-2490.
[5]	Yuyang SUN, Minjie ZHANG, Jie HU. Zero-shot dialogue state tracking domain transfer model based on semantic prefix-tuning [J]. Journal of Computer Applications, 2025, 45(7): 2221-2228.
[6]	Jin XIE, Surong CHU, Yan QIANG, Juanjuan ZHAO, Hua ZHANG, Yong GAO. Dual-branch distribution consistency contrastive learning model for hard negative sample identification in chest X-rays [J]. Journal of Computer Applications, 2025, 45(7): 2369-2377.
[7]	Dawei YANG, Xihai XU, Wei SONG. Relation extraction method combining semantic enhancement and perception attention [J]. Journal of Computer Applications, 2025, 45(6): 1801-1808.
[8]	Haijie WANG, Guangxin ZHANG, Hai SHI, Shu CHEN. Document-level relation extraction based on entity representation enhancement [J]. Journal of Computer Applications, 2025, 45(6): 1809-1816.
[9]	Wenjing YAN, Ruidong WANG, Min ZUO, Qingchuan ZHANG. Recipe recommendation model based on hierarchical learning of flavor embedding heterogeneous graph [J]. Journal of Computer Applications, 2025, 45(6): 1869-1878.
[10]	Jie HU, Cui WU, Jun SUN, Yan ZHANG. Document-level relation extraction model based on anaphora and logical reasoning [J]. Journal of Computer Applications, 2025, 45(5): 1496-1503.
[11]	Qingli CHEN, Yuanbo GUO, Chen FANG. Clustering federated learning algorithm for heterogeneous data [J]. Journal of Computer Applications, 2025, 45(4): 1086-1094.
[12]	Renjie TIAN, Mingli JING, Long JIAO, Fei WANG. Recommendation algorithm of graph contrastive learning based on hybrid negative sampling [J]. Journal of Computer Applications, 2025, 45(4): 1053-1060.
[13]	Yu WANG, Xianjin FANG, Gaoming YANG, Yifeng DING, Xinlu YANG. Active defense against face forgery based on attention mask and feature extraction [J]. Journal of Computer Applications, 2025, 45(3): 904-910.
[14]	Qiurun HE, Jie HU, Bo PENG, Tianyuan LI. Fabric defect detection algorithm based on context information and multi-scale feature fusion [J]. Journal of Computer Applications, 2025, 45(2): 640-646.
[15]	Tianqi ZHANG, Shuang TAN, Xiwen SHEN, Juan TANG. Image watermarking method combining attention mechanism and multi-scale feature [J]. Journal of Computer Applications, 2025, 45(2): 616-623.