The DOI matches its Crossref record; 51 of the 51 references it cites were found in a registry or an index.
Learning Text Similarity with Siamese Recurrent Networks
Open accessAbstract
This paper presents a deep architecture for learning a similarity metric on variablelength character sequences.The model combines a stack of character-level bidirectional LSTM's with a Siamese architecture.It learns to project variablelength strings into a fixed-dimensional embedding space by using only information about the similarity between pairs of strings.This model is applied to the task of job title normalization based on a manually annotated taxonomy.A small data set is incrementally expanded and augmented with new sources of variance.The model learns a representation that is selective to differences in the input that reflect semantic differences (e.g., "Java developer" vs. "HR manager") but also invariant to nonsemantic string differences (e.g., "Java developer" vs. "Java programmer").
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Siamese networks are widely applied in nonlinear metric learning across domains such as computer vision and speech recognition.”
From Reducing Label Dependency in Human Activity Recognition with Wearables: From Supervised Learning to Novel Weakly Self-Supervised Approaches (2025) · cites this paper as [47]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Hochreiter, S., & Schmidhuber, J. (1997) · Neural Computation · doi:10.1162/neco.1997.9.8.1735 · Cited by 102,318 · cited 2×
“In this equation and below the logistic function is denoted by σ The Long Short-Term Memory (Hochreiter and Schmidhuber, 1997) variant of RNNs in particular has had success in tasks related to natural language processing, such as text classification (Graves, 2012) and language translation (Sutskever…”
MethodAdam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946 · cited 1×
“The parameters of the model are optimized using the Adam method (Kingma and Ba, 2014) and each model is trained until convergence.”
MethodGlove: Global Vectors for Word Representation
Pennington, J., Socher, R., & Manning, C. (2014) · Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.3115/v1/d14-1162 · Cited by 34,071 · cited 2×
“, 2013) and Glove (Pennington et al., 2014) have seen much use in such models, but some go beyond the word level and represent text as a sequence of characters (Kim et”
Background(2014) · Cited by 33,419
Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation
Cho, K., van Merrienboer, B., Gulcehre, C., Bahdanau, D., Bougares, F., Schwenk, H., & Bengio, Y. (2014) · Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.3115/v1/d14-1179 · Cited by 25,545 · cited 2×
“In the last few years, neural network models have been applied to tasks ranging from machine translation (Zou et al., 2013; Cho et al., 2014) to question answering (Weston et al.”
MethodDistributed Representations of Words and Phrases and their Compositionality
Mikolov, T., Sutskever, I., Chen, K., Corrado, G., & Dean, J. (2013) · arXiv · doi:10.48550/arxiv.1310.4546 · Cited by 24,533 · cited 3×
“For example, word similarity models based on dense embeddings (Mikolov et al., 2013) have recently been applied in diverse settings, such as sentiment analysis (dos Santos and Gatti, 2014) and recommender systems (Barkan and Koenig-stein, 2016).”
BackgroundMethodReducing the dimensionality of data with neural networks
Hinton, G. E., & Salakhutdinov, R. R. (2006) · Science · doi:10.1126/science.1127647 · Cited by 20,928
Sequence to Sequence Learning with Neural Networks
Sutskever, I., Vinyals, O., & Le, Q. V. (2014) · arXiv · doi:10.48550/arxiv.1409.3215 · Cited by 17,766 · cited 1×
“…and below the logistic function is denoted by σ The Long Short-Term Memory (Hochreiter and Schmidhuber, 1997) variant of RNNs in particular has had success in tasks related to natural language processing, such as text classification (Graves, 2012) and language translation (Sutskever et al., 2014).”
BackgroundBidirectional recurrent neural networks
Schuster, M., & Paliwal, K. K. (1997) · IEEE Transactions on Signal Processing · doi:10.1109/78.650093 · Cited by 10,172 · cited 1×
“(1) Bidirectional RNNs (Schuster and Paliwal, 1997) incorporate both future and past context by running the reverse of the input through a separate RNN.”
BackgroundNatural Language Processing (almost) from ScratchPreprint
Collobert, R., Weston, J., Bottou, L., Karlen, M., Kavukcuoglu, K., & Kuksa, P. P. (2011) · arXiv (Cornell University) · doi:10.48550/arxiv.1103.0398 · Cited by 6,923 · cited 1×
“The use of (deep) neural networks for NLP has recently received much attention, starting from the seminal papers employing convolutional networks on traditional NLP tasks (Collobert et al., 2011) and the availability of high quality semantic word representations (Mikolov et al., 2013).”
Method
Research works printed but not linked: 5
- Jane Bromley, James W Bentz, Léon Bottou, Is- abelle Guyon, Yann LeCun, Cliff Moore, Eduard Säckinger, and Roopak Shah. 1993. Signature ver- ification using a "siamese" time delay neural net- work. International Journal of Pattern Recognition and Artificial Intelligence, 7(04):669-688.
- Tomas Mikolov, Ilya Sutskever, Kai Chen, Greg S Cor- rado, and Jeff Dean. 2013. Distributed representa- tions of words and phrases and their compositional- ity. In Advances in neural information processing systems, pages 3111-3119.
- Mike Schuster and Kuldip K Paliwal. 1997. Bidirec- tional recurrent neural networks. Signal Processing, IEEE Transactions on, 45(11):2673-2681.
- Martijn Spitters, Remko Bonnema, Mihai Rotaru, and Jakub Zavrel. 2010. Bootstrapping information ex- traction mappings by similarity-based reuse of tax- onomies. In CEUR Workshop Proceedings, volume 673.
- Longqi Yang, Yin Cui, Fan Zhang, John P Pollak, Serge Belongie, and Deborah Estrin. 2015. Plate- click: Bootstrapping food preferences through an adaptive visual interface. In Proceedings of the 24th ACM International on Conference on Informa- tion and Knowledge Management, pages 183-192. ACM.
Cited by
Sorted by citations
Items 101–200 of 341
Prabhakar, S. K., Rajaguru, H., & Won, D.-O. (2021) · Scientific Programming · doi:10.1155/2021/2420254 · Cited by 12
“For the text matching applications, Siamese Neural Networks (SNNs) were used widely [33].”
MethodMatching Descriptions to Spatial Entities Using a Siamese Hierarchical Attention Network
Ma, K., Wu, L., Tao, L., Li, W., & Xie, Z. (2018) · IEEE Access · doi:10.1109/access.2018.2837666 · Cited by 12
“Following [36], at the output layer of the model, the cosine distance is employed to measure the similarity of p and e.”
MethodEntity Synonym Discovery via Multipiece Bilateral Context Matching
Zhang, C., Li, Y., Du, N., Fan, W., & Yu, P. S. (2020) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2020/199 · Cited by 11
“The following loss function L Siamese is used in training for the Siamese architecture [Neculoiu et al. , 2016]: where it contains losses for two cases: L + ( e; k ) when e and k are synonyms to each other ( y = 1 ), and L (cid:0) ( e; k ) when e and k are not ( y = 0 ): where s ( (cid:1) ) is a…”
BackgroundMethodInfluentialStacking-BERT model for Chinese medical procedure entity normalization
Li, L., Zhai, Y., Gao, J., Wang, L., Hou, L., & Zhao, J. (2022) · Mathematical Biosciences & Engineering · doi:10.3934/mbe.2023047 · Cited by 12
Recovering Question Answering Errors via Query Revision
Yavuz, S., Gür, İ., Su, Y., & Yan, X. (2017) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/d17-1094 · Cited by 11
“The effectiveness of these architectures has been shown in other studies (Neculoiu et al., 2016; Hermann et al., 2015; Chen et al., 2016; Mueller and Thyagarajan, 2016) as well.”
BackgroundDual Neural Personalized Ranking
Kim, S., Lee, J., & Shim, H. (2019) · International World Wide Web Conference (WWW) · doi:10.1145/3308558.3313585 · Cited by 11
“This approach is also used in other research domains such as computer vision [29, 30], information retrieval [11] and, natural language processing [3, 17].”
MethodCost–effective Variational Active Entity Resolution
Bogatu, A., Paton, N. W., Douthwaite, M., Davie, S. J., & Freitas, A. (2021) · IEEE International Conference on Data Engineering (ICDE) · doi:10.1109/icde51399.2021.00114 · Cited by 13
“, a class of neural network that contain two or more identical sub–networks) [8], [36].”
BackgroundMethodMuneer, I., & Nawab, R. M. A. (2021) · ACM Transactions on Asian and Low-Resource Language Information Processing · doi:10.1145/3473331 · Cited by 12
“Siamese recurrent architecture has been used for a range of applications, including learning text similarities [45], visual tracking [67], homoglyph attacks detection [64], human re-identification [62], and so on.”
BackgroundInfluentialSIMPA: Statement-to-Item Matching Personality Assessment from text
Gjurković, M., Vukojević, I., & Šnajder, J. (2021) · Future Generation Computer Systems · doi:10.1016/j.future.2021.12.014 · Cited by 12
Arabic Medical Community Question Answering Using ON-LSTM and CNN
Balla, H. A. M. N., Salvador, M. L., & Delany, S. J. (2022) · International Conference on Machine Learning and Computing (ICMLC) · doi:10.1145/3529836.3529913 · Cited by 13
“This approach delivers higher scalability, in addition, it solves the spelling error problems in the sentences [30].”
BackgroundMethodTwo Stage Job Title Identification System for Online Job Advertisements
Rahhal, I., Carley, K. M., Kassou, I., & Ghogho, M. (2023) · IEEE Access · doi:10.1109/access.2023.3247866 · Cited by 17
“Moreover, in [16], they leveraged string similarity, where similar job titles were fed to the siamese network to learn to classify job titles.”
BackgroundMethodDisambiguation of company names via deep recurrent networks
Basile, A., Crupi, R., Grasso, M., Mercanti, A., Regoli, D., Scarsi, S., Yang, S., & Cosentini, A. (2023) · Expert Systems with Applications · doi:10.1016/j.eswa.2023.122035 · Cited by 10
“Neculoiu et al. (2016) propose a Siamese Deep Learning model as well, with the slightly different task of mapping strings to a predefined set of job names.”
MethodUnsupervised Contextual Paraphrase Generation using Lexical Control and Reinforcement LearningPreprint
Garg, S., Prabhu, S., Misra, H., & Srinivasaraghavan, G. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2103.12777 · Cited by 14
“To assess textual entailment, we train a model to generate a joint space embedding for conversation contexts and agent responses by using a ranking based loss (Neculoiu et al., 2016; Reimers and Gurevych, 2019a).”
MethodSimilarity matching of medical question based on Siamese network
Li, Q., & He, S. (2023) · BMC Medical Informatics and Decision Making · doi:10.1186/s12911-023-02161-z · Cited by 12
“Neculoiu [27] et al. also relied on Siamese networks, using a BiLSTM network based on character-level embedding to process sentence pairs, extracting semantic features of sentences that can learn semantic differences and semantic invariance among different words, then finally perform calculations…”
MethodFew-shot entity linking of food names
Feher, D., Ibrahim, F., Cheng, Z., Schlegel, V., Maidment, T., Bagshaw, J., & Batista-Navarro, R. (2023) · Information Processing & Management · doi:10.1016/j.ipm.2023.103463 · Cited by 10
Xu, L., Ma, H., Guan, Y., Liu, J., Huang, H., Zhang, Y., & Tian, L. (2023) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2023.3304974 · Cited by 9
Siamese Multiplicative LSTM for Semantic Text Similarity
Lv, C., Wang, F., Wang, J., Yao, L., & Du, X. (2020) · International Conference on Algorithms, Computing and Artificial Intelligence (ACAI) · doi:10.1145/3446132.3446160 · Cited by 9
“[25] proposed a character level two layer bidirectional LSTM network to evaluate the similarity between job titles by using the contrastive loss function.”
BackgroundMethodImproved Chinese Sentence Semantic Similarity Calculation Method Based on Multi-Feature Fusion
Liu, L., Wang, Q., & Li, Y. (2021) · Journal of Advanced Computational Intelligence and Intelligent Informatics · doi:10.20965/jaciii.2021.p0442 · Cited by 9
Tang, L., & Clematide, S. (2021) · Natural Legal Language Processing Workshop (NLLP) · doi:10.18653/v1/2021.nllp-1.12 · Cited by 9
“In the best case, such semantic similarity models at paragraph level should be invariant to differences in the input that do not matter for approximating relevance in this task, while they will be more selective to differences that do matter (Neculoiu et al. 2016).”
BackgroundMeasuring employer attractiveness in diverse talent markets
Li, L., Lappas, T., & Liu, R. (2023) · Decision Support Systems · doi:10.1016/j.dss.2023.114079 · Cited by 8
Learning Relational Representations by Analogy using Hierarchical Siamese Networks
Rossiello, G., Gliozzo, A., Farrell, R. G., Fauceglia, N. R., & Glass, M. R. (2018) · North American Chapter of the Association for Computational Linguistics · doi:10.18653/v1/n19-1327 · Cited by 14
“This kind of neural network has been used in both computer vision [Koch et al., 2015] and natural language processing [Mueller and Thyagarajan, 2016, Neculoiu et al., 2016] in order to map two similar instances close in a feature space.”
MethodA bidirectional Siamese recurrent neural network for accurate gait recognition using body landmarks
Progga, P. H., Rahman, M. J., Biswas, S., Ahmed, S., Anwary, A. R., & Shatabda, S. (2024) · Neurocomputing · doi:10.1016/j.neucom.2024.128313 · Cited by 9
Textual Entailment in Legal Bar Exam Question Answering Using Deep Siamese Networks
Kim, M., Lu, Y., & Goebel, R. G. (2018) · Lecture notes in computer science · doi:10.1007/978-3-319-93794-6_3 · Cited by 8
Matching Recruiters and Jobseekers on Twitter
Khatua, A., & Nejdl, W. (2020) · IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM) · doi:10.1109/asonam49781.2020.9381392 · Cited by 10
“Existing literature has obtained the state of the art results by using CNN [4] and RNN [5], [11] for the above sentence similarity task.”
MethodLearning Neural Textual Representations for Citation Recommendation
Kieu, B. T., Unanue, I. J., Pham, S. B., Phan, X.-H., & Piccardi, M. (2020) · UTS ePRESS (University of Technology Sydney) · Cited by 8
Li, Y., Yu, X., & Koudas, N. (2021) · Proceedings of the VLDB Endowment · doi:10.14778/3476249.3476263 · Cited by 9
“Siamese networks have been successfully utilized in deep metric learning tasks [50, 52] in computer vision, capturing both intra-class similarity and inter-class discrimination in many challenging tasks including face recognition.”
BackgroundA Neural Network Architecture for Program Understanding Inspired by Human Behaviors
Zhu, R., Yuan, L., Li, X., Gao, M., & Cai, W. (2022) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2022.acl-long.353 · Cited by 8
“In the code clone detection task, as suggested by (Neculoiu et al., 2016), we double the PGNN-EK to a siamese neural network to calculate code similarity.”
MethodOff-Topic Spoken Response Detection with Word Embeddings
Yoon, S., Lee, C., Choi, I., Wang, X., Mulholland, M., & Evanini, K. (2017) · Interspeech (USB) · doi:10.21437/interspeech.2017-388 · Cited by 8
“It has been applied to the detection of similar sentences and has shown competitive performance [9, 10, 18].”
MethodIntelligent Voice Instructor-assistant System for Collaborative and Interactive Classes
Baker, M., Hu, X., Luca, G. D., & Chen, Y. (2021) · Journal of Artificial Intelligence and Technology · doi:10.37965/jait.2021.0003 · Cited by 8
“The Siamese LSTM layers are designed based on LSTM cell [24], [25].”
Al-Quran recitation verification for memorization test using Siamese LSTM network
Rajagede, R. A., & Hastuti, R. P. (2021) · Communications in Science and Technology · doi:10.21924/cst.6.1.2021.344 · Cited by 14
“Combining Siamese with LSTM network is done previously to handle data in sequences such as speech or text data [11,15].”
MethodZola, P., Cortez, P. C., & Brentari, E. (2020) · Neural Computing and Applications · doi:10.1007/s00521-020-04991-8 · Cited by 8
“, Word2Vec) to get a numerical representation of the texts [27, 32].”
BackgroundIntegrating legal event and context information for Chinese similar case analysis
Dan, J., Xu, L., & Wang, Y. (2023) · Artificial Intelligence and Law · doi:10.1007/s10506-023-09377-4 · Cited by 7
“Inspired by the Siamese network (Neculoiu et al. 2016), we design our encoder based on a shared-weight BERT to encode every paragraph, which is beneficial to reducing model parameters while fully considering the interaction information between different documents.”
Word-Embedding based Content Features for Automated Oral Proficiency Scoring
Yoon, S., Loukina, A., Lee, C., Mulholland, M., Wang, X., & Choi, I. (2018) · International Conference on Computational Linguistics · Cited by 7
Semantic-Preserving Abstractive Text Summarization with Siamese Generative Adversarial Net
Sheng, X., Xu, L., Xu, Y., Jiang, D., & Ren, B. (2022) · Findings of the Association for Computational Linguistics: NAACL · doi:10.18653/v1/2022.findings-naacl.163 · Cited by 6
“(Neculoiu et al., 2016) considers similarity matching of a sentence pair as a binary classification task and replaces the Manhattan distance with cosine similarity.”
BackgroundArtene, C.-G., Oprişa, C., Buţincu, C. N., & Leon, F. (2023) · Mathematics · doi:10.3390/math11092053 · Cited by 8
Su, P.-C., Huang, B.-H., & Kuo, T.-Y. (2024) · Electronics · doi:10.3390/electronics13040804 · Cited by 8
“[44] utilized a Siamese network for feature extraction, followed by distinct classification and regression This arrangement is commonly referred to as Siamese-RPN.”
MethodCODE-SMASH: Source-Code Vulnerability Detection Using Siamese and Multi-Level Neural Architecture
Han, S., Nam, H., Kang, J., Kim, K., Cho, S., & Lee, S. (2024) · IEEE Access · doi:10.1109/access.2024.3432323 · Cited by 7
“The core of CODE-SMASH is based not only on a Siamese architecture, known for its effectiveness in comparing the similarity between two source codes [42], [44], but also on a hierarchical feature encoder designed to extract features across various levels of a source code.”
BackgroundContrastive Late Fusion for 3D Object Detection
Zhang, T., Liang, Z., Yang, Y., Yang, X., Zhu, Y., & Wang, J. (2024) · IEEE Transactions on Intelligent Vehicles · doi:10.1109/tiv.2024.3454085 · Cited by 7
Agree or Disagree: Predicting Judgments on Nuanced Assertions
Wojatzki, M., Zesch, T., Mohammad, S. M., & Kiritchenko, S. (2018) · Joint Conference on Lexical and Computational Semantics · doi:10.18653/v1/s18-2026 · Cited by 7
“SNNs have been successfully used to predict text similarity (Mueller and Thyagarajan, 2016; Neculoiu et al., 2016) and match pairs of sentences (e.”
MethodRevisiting paraphrase question generator using pairwise discriminator
Patro, B. N., Chauhan, D., Kurmi, V. K., & Namboodiri, V. P. (2020) · Neurocomputing · doi:10.1016/j.neucom.2020.08.022 · Cited by 7
Triplet-Trained Vector Space and Sieve-Based Search Improve Biomedical Concept Normalization
Xu, D., & Bethard, S. J. (2021) · Workshop on Biomedical Language Processing (BioNLP) · doi:10.18653/v1/2021.bionlp-1.2 · Cited by 8
Transformer-based Denoising Adversarial Variational Entity Resolution
Li, S., & Wu, H. (2023) · Journal of Intelligent Information Systems · doi:10.1007/s10844-022-00773-x · Cited by 7
“Inspired by (Bogatu et al., 2021), to optimize both objectives simultaneously, we adopt a contrastive loss function (Neculoiu et al., 2016), as shown in Eq.”
MethodMishra, R., & Zhang, S. (2021) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3459637.3482376 · Cited by 7
“The sentence matching and lexical similarity based methods [18] are not a good fit for headline incongruence problem due to inherent challenges such as relative length and vocabulary mismatch between the news headline text and its body content.”
BackgroundBayesian Optimization for Selecting Efficient Machine Learning ModelsPreprint
Wang, L., Dernoncourt, F., & Bui, T. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2008.00386 · Cited by 7
“…which is a standard technique widely applied for this task for balancing speed and performance; deep neural networks for the insurance domain, which consists of a set of BiLSTM models for capturing user search terms and answer candidates, respectively, and computing their similarity score [7].”
MethodDEEP LEARNING FOR SEMANTIC MATCHING: A SURVEY
Li, H., Govind, Y., Mudgal, S., Rekatsinas, T., & Doan, A. (2021) · Journal of Computer Science and Cybernetics · doi:10.15625/1813-9663/37/4/16151 · Cited by 7
“Many works have applied DL to this problem [20, 26, 49, 50, 73, 75, 79, 100, 120].”
BackgroundMethodInfluentialSimilarity learning networks uniquely identify individuals of four marine and terrestrial species
Kabuga, E., Langley, I., Civil, M. A., Measey, J. G., Bah, B., & Durbach, I. N. (2024) · Ecosphere · doi:10.1002/ecs2.70012 · Cited by 6
MGMSN: Multi-Granularity Matching Model Based on Siamese Neural Network
Wang, X., & Yang, H. (2022) · Frontiers in Bioengineering and Biotechnology · doi:10.3389/fbioe.2022.839586 · Cited by 6
“…mainstream text matching algorithms, they are Deep Structured Semantic Model (DSSM) (Huang et al. (2013)), Siamese LSTM network (Siamese LSTM) (Neculoiu et al. (2016)), Attention-Based Convolutional Neural Network for Modeling Sentence Pairs (ABCNN) (Yin et al. (2016)), Enhance Sequential…”
MethodDu, H., Huang, Z., & Zhang, Y. (2024) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2024.3379567 · Cited by 6
Neural Scoring of Logical Inferences from Data using Feedback.
Susaiyah, A., Härmä, A., Reiter, E., & Petković, M. (2021) · International Journal of Interactive Multimedia and Artificial Intelligence · doi:10.9781/ijimai.2021.02.004 · Cited by 7
“Based on previous works on similar input/ output constraints [27], [28], we came up with three neural network architectures, namely, a recurrent neural network (RNNA), a modified RNN (RNNB) and a siamese network (SIAM).”
MethodEvaluating SIAMESE Architecture Neural Models for Arabic Textual Similarity and Plagiarism Detection
Aliane, A. A., & Aliane, H. (2020) · International Symposium Informatics and its Applications · doi:10.1109/isia51297.2020.9416550 · Cited by 6
“[14] used a combination of CNN and LSTM. [15] used a deep architecture for learning a similarity metric on variable length character sequences.”
MethodMultimodal Product Matching and Category Mapping: Text+Image based Deep Neural Network
Gupte, K., Pang, L. X., Vuyyuri, H., & Pasumarty, S. (2021) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata52589.2021.9671384 · Cited by 7
“A more useful approach is siamese network [2] [7] [8], which allows us to train a similarity function between embed-dings.”
BackgroundSimilarity Calculation Method of Siamese-CNN Judgment Document Based on TinyBERT
Wan, Z., & Ye, N. (2021) · International Conference on Intelligent Computing, Automation and Applications (ICAA) · doi:10.1109/icaa53760.2021.00013 · Cited by 6
Edge Weight Updating Neural Network for Named Entity Normalization
Jeon, S. H., & Cho, S. (2022) · Neural Processing Letters · doi:10.1007/s11063-022-11102-2 · Cited by 7
“Siamese RNN model successfully apprehends the morphological similarity between strings [35].”
BackgroundBocharova, M. Y., Malakhov, E. V., & Mezhuyev, V. I. (2023) · Applied Aspects of Information Technology · doi:10.15276/aait.06.2023.4 · Cited by 13
“ISSN 2663-7723 (Online) In [1] use a large hand-labeled dataset of similar job titles and train a Siamese Recurrent Neural network to project job titles representation into the vector space.”
BackgroundMethodVariational model for low-resource natural language generation in spoken dialogue systems
Tran, V.-K., & Nguyen, L.-M. (2020) · Computer Speech & Language · doi:10.1016/j.csl.2020.101120 · Cited by 6
Energy-based Self-attentive Learning of Abstractive Communities for Spoken Language UnderstandingPreprint
Shang, G., Tixier, A. J. ‐., Vazirgiannis, M., & Lorré, J.-P. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1904.09491 · Cited by 6
“It performed better than the traditional contrastive loss (Chopra et al., 2005; Neculoiu et al., 2016) in early experiments.”
BackgroundSheng, T., & Huber, M. (2025) · Sensors · doi:10.3390/s25134032 · Cited by 5
“In natural language processing, Mueller and Thyagarajan [46] introduced a Siamese RNN to evaluate semantic similarity between sentences, while Neculoiu et al. [47] employed one for job title normalization in recruitment analysis.”
Deep Learning in Stance Detection: A Survey
Gera, P., & Neal, T. J. (2025) · ACM Computing Surveys · doi:10.1145/3744641 · Cited by 7
“Siamese networks, which use twin subnetworks with shared weights (e.g., LSTMs) to compare text representations [139], have shown promise for text classiication.”
Quality Estimation for Automatically Generated Titles of eCommerce Browse Pages
Ueffing, N., de Souza, J. G. C., & Leusch, G. (2018) · Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT) · doi:10.18653/v1/n18-3007 · Cited by 5
“X = {(xi , xi , yi)}i=1 is defined following (Neculoiu et al., 2016):”
BackgroundJeon, S. H., Lee, H. J., Park, J., & Cho, S. (2023) · Intelligent Data Analysis · doi:10.3233/ida-227129 · Cited by 5
“The morphological similarity between strings was linked by the Siamese RNN model [29].”
Koochaki, F., & Najafizadeh, L. (2024) · IEEE Transactions on Neural Systems and Rehabilitation Engineering · doi:10.1109/tnsre.2024.3391067 · Cited by 6
“SNN is suitable for applications with small datasets [28], [29], and tasks in which comparing and classifying pairs of data is crucial, and has been employed in various applications such as diagnosing spinal metastases [30], detection of diabetic retinopathy in retinal fundus photographs [31],…”
MethodMultimodal learning for non-small cell lung cancer prognosis
Wu, Y., Wang, Y., Huang, X., Wang, H., Yang, F., Sun, W., Ling, S. H., & Su, S. W. (2025) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2025.107663 · Cited by 6
Fan, J., Nichols, E., Tompkins, D. G., Méndez, A. E. M., Elizalde, B., & Pasquier, P. (2020) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp40776.2020.9053972 · Cited by 5
“images and video [12-15].”
Peng, D., Hao, B., Tang, X., Chen, Y., Sun, J., & Wang, R. (2020) · International Conference on Control, Robotics and Intelligent System (CCRIS) · doi:10.1145/3437802.3437806 · Cited by 5
“Inspired by [31]], we calculate the similarity between text representation vectors through the Siamese network structure.”
MethodIncorporating Term Definitions for Taxonomic Relation Identification
Sheng, Y., Wu, T., & Wang, X. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-41407-8_1 · Cited by 5
Investigating Citation Linkage as a Sentence Similarity Measurement Task Using Deep Learning
Roy, S. S., Mercer, R. E., & Urra, F. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-47358-7_50 · Cited by 5
“[15] investigated different neural network based techniques for generating fixed length vector representations of the sentences utilizing the siamese structure [46].”
BackgroundMethodInfluentialA Siamese CNN Architecture for Learning Chinese Sentence Similarity
Shi, H., Wang, C., & Sakai, T. (2020) · Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the International Joint Conference on Natural Language Processing: Student Research Workshop (AACL-IJCNLP SRW) · doi:10.18653/v1/2020.aacl-srw.4 · Cited by 5
“Subsequently, a Siamese Bi-LSTM structure was proposed in order to improve the result of sentence similarity (Neculoiu et al., 2016).”
BackgroundMethodResultInfluentialA New Semantic Approach to Improve Webpage Segmentation
Ghaemmaghami, S. S. S., & Miller, J. (2021) · Journal of Web Engineering · doi:10.13052/jwe1540-9589.2042 · Cited by 5
Muneer, I., & Nawab, R. M. A. (2022) · Language Resources and Evaluation · doi:10.1007/s10579-022-09613-4 · Cited by 5
Medical Answer Selection Based on Two Attention Mechanisms with BiRNN
Ma, J., Che, C., & Zhang, Q. (2018) · MATEC Web of Conferences · doi:10.1051/matecconf/201817601024 · Cited by 5
“Most papers (Mnih et al., 2014; Wang and Nyberg, 2015; Mueller and Thyagarajan, 2016; Neculoiu et al., 2016; Tan et al., 2016) are modeled by BiLSTM.”
Chinese Social Media Entity Linking Based on Effective Context with Topic Semantics
Ma, C., Sha, Y., Tan, J., Guo, L., & Peng, H. (2019) · IEEE Annual International Computer Software and Applications Conference (COMPSAC) · doi:10.1109/compsac.2019.00063 · Cited by 5
“The LSTM is superior to basic RNNs for learning long range dependencies through its use of memory cell units that can store access information across lengthy input sequences [27].”
BackgroundMethodHuman activity recognition with deep metric learners.
Martin, K., Wijekoon, A., & Wiratunga, N. (2020) · Open Access Institutional Repository at Robert Gordon University (Robert Gordon University) · Cited by 5
Virtual Adversarial Training for Semi-supervised Verification Tasks
Noroozi, V., Bahaadini, S., Zheng, L., Xie, S., & Yu, P. S. (2019) · European Signal Processing Conference (EUSIPCO) · doi:10.23919/eusipco.2019.8902573 · Cited by 5
“Siamese networks are widely used in similarity learning [3], [4], [13], embedding learning [14], [15], verification [2], [16], [17], and retrieval [18].”
Backgroundde Souza, J. V. A., Oliveira, L. E. S. E., Gumiel, Y. B., Carvalho, D. R., & Moro, C. M. C. (2019) · Cited by 5
ERCNN: Enhanced Recurrent Convolutional Neural Networks for Learning Sentence Similarity
Xie, N., Li, S., & Zhao, J. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-32381-3_10 · Cited by 4
Normalization of Web of Science Institution Names Based on Deep Learning
Jia, Z., Fang, Z., & Zhang, H. (2024) · Algorithms · doi:10.3390/a17070312 · Cited by 4
“SRN [29]: SRN is a character-level model that encodes entities as a sequence of characters using BiLSTM.”
MethodJoint Multi-field Siamese Recurrent Neural Network for Entity Resolution
Lv, Y., Qi, L., Huo, J., Wang, H., & Gao, Y. (2018) · Lecture notes in computer science · doi:10.1007/978-3-319-97310-4_55 · Cited by 4
Wolcott, L., Clements, W., & Saripalli, P. (2018) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata.2018.8622516 · Cited by 4
“There are even some instances of deep learning being applied to matching problems ([10], [11]), and Siamese networks used for single text phrase or image similarity comparison ([12], [13]), but not in the context of scalability and big data.”
BackgroundDeep Neural Query Understanding System at Expedia Group
Chandrasekaran, R., Pathak, H. N., & Yano, T. (2020) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata50022.2020.9378495 · Cited by 4
“Cross entropy loss was used for intent classifier, CRF loss for NER [4], pairwise contrastive loss for siamese network [25] and Mean Squared Error for Re-ranker.”
BackgroundMethodCapsTM: capsule network for Chinese medical text matching
Yu, X., Shen, Y., Ni, Y., Huang, X., Wang, X., Chen, Q., & Tang, B. (2021) · BMC Medical Informatics and Decision Making · doi:10.1186/s12911-021-01442-9 · Cited by 4
“Most of deep neural networks are based on Siamese network [9] which aims to represent two pieces of text by the same structure.”
BackgroundDuffner, S., García, C., Idrissi, K., & Baskurt, A. M. (2012) · Springer eBooks · doi:10.1007/978-3-030-74478-6_5 · Cited by 4
“[ NVR16] uses a Bidirectional LSTM to compute similarities between words.”
Person Re-Identification by Low-Dimensional Features and Metric Learning
Chen, X., Xu, H., Li, Y., & Bian, M. (2021) · Future Internet · doi:10.3390/fi13110289 · Cited by 4
“In [26], the authors proposed a novel Siamese LSTM architecture that can process image regions sequentially and enhance the discriminative capability of the local feature representation by leveraging contextual information.”
BackgroundDeep Unified Representation for Heterogeneous Recommendation
Lu, C., Yin, M., Shen, S., Ji, L., Liu, Q., & Yang, H. (2022) · International World Wide Web Conference (WWW) · doi:10.1145/3485447.3512087 · Cited by 5
“The overall framework of our model is a basic twotower architecture that was first proposed for natural language tasks [18, 33] and then becomes a popular approach in large-scale recommendation system [57] because of its high efficiency.”
BackgroundLSGC: An Interactive Text Matching Model Combined with Enhanced Encoding
Wang, C., Xu, X., Wei, Z., Zhang, W., & Liu, Z. (2022) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn55064.2022.9889792 · Cited by 4
“The first type of framework relies on representation, where two texts are processed separately, then encoded by a Siamese network with shared parameters to obtain representations, and the final similarity of the sentence pair is calculated by a specific function [4, 5].”
BackgroundMethodLiu, P., Ye, W., Duan, H., Li, X., Zhang, S., Yao, C., & Li, Y. (2023) · Cybersecurity · doi:10.1186/s42400-023-00160-1 · Cited by 4
“Siamese networks are twin networks used for metric learning, i.e., they are composed of two identical subnetworks sharing the same weights (Neculoiu et al. 2016).”
BackgroundImproving Word Embeddings through Iterative Refinement of Word- and Character-level Models
Ha, P., Zhang, S., Djuric, N., & Vučetić, S. (2020) · International Conference on Computational Linguistics · doi:10.18653/v1/2020.coling-main.104 · Cited by 4
“We note that this approach does not require manually labeled training data, unlike the method proposed in (Neculoiu et al., 2016).”
MethodBeyond Topics: Discovering Latent Healthcare Objectives from Event Sequences
Caruana, A., Bandara, M., Catchpoole, D. R., & Kennedy, P. J. (2022) · Lecture notes in computer science · doi:10.1007/978-3-030-97546-3_30 · Cited by 4
“They been used for text similarity and sentence embedding in NLP [18,22].”
MethodDeep Learning for Text Matching: A Survey
Huang, Z., & Cao, L. (2021) · International Conference on Data Science and Business Analytics (ICDSBA) · doi:10.1109/icdsba53075.2021.00022 · Cited by 4
Zero-Shot Question Classification Using Synthetic Samples
Fu, H., Yuan, C., Wang, X., Sang, Z., Hu, S., & Shi, Y. (2018) · IEEE International Conference on Cloud Computing and Intelligence Systems (CCIS) · doi:10.1109/ccis.2018.8691209 · Cited by 4
Gaussian mixture embedding of multiple node roles in networks
Chen, Y., Pu, J., Liu, X., & Zhang, X. (2019) · World Wide Web · doi:10.1007/s11280-019-00743-4 · Cited by 4
“Our objective function setting and optimization are relevant to energy-based learning, which has been actively used for unsupervised representation learning [33], natural language processing [41], text embedding [42] and computer vision [8, 55].”
BackgroundEnergy-based Self-attentive Learning of Abstractive Communities for Spoken Language Understanding
Shang, G., Tixier, A. J. ‐., Vazirgiannis, M., & Lorré, J.-P. (2020) · Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the International Joint Conference on Natural Language Processing (AACL-IJCNLP) · doi:10.18653/v1/2020.aacl-main.34 · Cited by 12
“It performed better than the traditional contrastive loss (Chopra et al., 2005; Neculoiu et al., 2016) in early experiments.”
BackgroundNSURL-2019 Shared Task 8: Semantic Question Similarity in ArabicPreprint
Seelawi, H., Mustafa, A., Al-Bataineh, H., Farhan, W., & Al-Natsheh, H. T. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1909.09691 · Cited by 8
Online Similarity Learning with Feedback for Invoice Line Item MatchingPreprint
Maurya, C. K., Gantayat, N., Dechu, S., & Horváth, T. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2001.00288 · Cited by 4
“• Siamese [15]: The method of [15] uses character level bidirectional LSTM combined with the Siamese architecture to learn a fixed dimensional representation from variable length sequences using only information about the similarity between pairs of strings.”
InfluentialSource Code Plagiarism Detection Using Siamese BLSTM Network and Embedding Models
Manahi, M., Sulaiman, S., & Bakar, N. S. A. A. (2022) · Lecture Notes in Electrical Engineering · doi:10.1007/978-981-16-8515-6_31 · Cited by 3
Alian, M., Himdi, H., & Shaalan, K. F. (2025) · Arabian Journal for Science and Engineering · doi:10.1007/s13369-025-10734-5 · Cited by 3
The Use of Artificial Intelligence Models in the Automated Knowledge Assessment System
Katerynchuk, I., Komarnytska, O., & Балендр, А. (2024) · Lecture notes in electrical engineering · doi:10.1007/978-3-031-61221-3_13 · Cited by 4
Zero-Shot Cross-Lingual Named Entity Recognition via Progressive Multi-Teacher Distillation
Li, Z., Hu, C., Zhang, R., Chen, J., & Guo, X. (2024) · IEEE/ACM Transactions on Audio Speech and Language Processing · doi:10.1109/taslp.2024.3449029 · Cited by 3
“The siamese network is originally introduced to treat signature verification as a matching problem [49] and has been successfully applied to transfer learning such as one-shot image recognition [50] and text similarity [51].”
BackgroundA Hierarchical Information Compression Approach for Knowledge Discovery From Social Multimedia
Liu, Z., Weng, Y., Xu, R., Chaomurilige, C., & Gao, H. (2024) · IEEE Transactions on Computational Social Systems · doi:10.1109/tcss.2024.3440997 · Cited by 3
Han, S., Kim, M., Kang, J., Kim, K., Lee, S., & Lee, S. (2024) · IEEE Access · doi:10.1109/access.2024.3474857 · Cited by 3
“In particular, as the Siamese neural network [16] has emerged as a powerful architecture for text similarity tasks [28], [29], [30], [31], [32], [33], Siamese architecture-based methods for source code vulnerability detection have been proposed [11], [12], [13].”
BackgroundHe, Z. (2025) · Neurocomputing · doi:10.1016/j.neucom.2025.130355 · Cited by 4
Metadata
- Authors
- Neculoiu, P., Versteegh, M., & Rotaru, M.
- ORCID
- Versteegh, Maarten
- Venue
- Proceedings of the 1st Workshop on Representation Learning for NLP
- Published
- 2016
- Pages
- 148-157
- OpenAlex
- W2510940142
- Cited by
- 341
- References
- 51
- Citations by year
- 2026: 6 · 2025: 21 · 2024: 30 · 2023: 31 · 2022: 43 · 2021: 63 · 2020: 64 · 2019: 37 · 2018: 36 · 2017: 9
- FWCI
- 14.73
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- Locations
- Workshop on Representation Learning for NLP (RepL4NLP) (publishedVersion, cc-by)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Physical Sciences › Computer Science › Information Systems › Data Quality and Management
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Handwritten Text Recognition Techniques
- Keywords
- text similarity, Siamese network, bidirectional LSTM, character embeddings, similarity metric learning, taxonomy (general classification scheme), data augmentation, semantic representation, string embeddings
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
