LSTM neural networks for language modeling
Open accessAbstract
Neural networks have become increasingly popular for the task of language modeling.Whereas feed-forward networks only exploit a fixed context length to predict the next word of a sequence, conceptually, standard recurrent neural networks can take into account all of the predecessor words.On the other hand, it is well known that recurrent networks are difficult to train and therefore are unlikely to show the full potential of recurrent models.These problems are addressed by a the Long Short-Term Memory neural network architecture.In this work, we analyze this type of network on an English and a large French language modeling task.Experiments show improvements of about 8 % relative in perplexity over standard recurrent neural network LMs.In addition, we gain considerable improvements in WER on top of a state-of-the-art speech recognition system.
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
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?
The size of the kNN-LM datastore is almost equal to the number of training tokens because the context is nearly unique due to the large context window size in modern recurrent or self-attentional NLMs.
“The size of such a datastore is almost equal to the number of training tokens because the context c t is (nearly) unique due to the large context window size in modern recurrent (Sundermeyer et al., 2012) or self-attentional (Vaswani et al., 2017) NLMs.”
From Efficient Nearest Neighbor Language Models (2021) · cites this paper as Sundermeyer et al., 2012
The Introduction says recurrent language models can take predecessor words into account, supporting the cited point about their broad context. The Experimental results discuss longer input sequences but do not establish that contexts are nearly unique or that a kNN-LM datastore is almost as large as the training-token count. (LLM, conf=0.86)
Experimental — accuracy not yet measuredRNNs have achieved promising performance enhancements in speech recognition, language modeling, video processing, and other sequence labeling tasks.
“They have achieved promising performance enhancements in many technical applications, such as speech recognition [1], language modeling [2], video processing [3], and many other sequence labeling tasks [4].”
From Deep Recurrent Neural Networks for Human Activity Recognition (2017) · cites this paper as [2]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Hochreiter, S., & Schmidhuber, J. (1997) · Neural Computation · doi:10.1162/neco.1997.9.8.1735 · Cited by 102,318 · see citing works
Learning representations by back-propagating errors
Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (1986) · Nature · doi:10.1038/323533a0 · Cited by 31,892 · see citing works
Neural networks for pattern recognition
Bishop, C. (1994) · Choice Reviews Online · doi:10.5860/choice.31-5500 · Cited by 18,537 · see citing works
Neural Networks for Pattern Recognition
Bishop, C. M. (1995) · doi:10.1093/oso/9780198538493.001.0001 · Cited by 12,287 · see citing works
Elman, J. L. (1990) · Cognitive Science · doi:10.1207/s15516709cog1402_1 · Cited by 11,212 · see citing works
Learning long-term dependencies with gradient descent is difficult
Bengio, Y., Simard, P., & Frasconi, P. (1994) · IEEE Transactions on Neural Networks · doi:10.1109/72.279181 · Cited by 8,542 · see citing works
Framewise phoneme classification with bidirectional LSTM and other neural network architectures
Graves, A., & Schmidhuber, J. (2005) · Neural Networks · doi:10.1016/j.neunet.2005.06.042 · Cited by 5,882 · see citing works
Recurrent neural network based language model
Mikolov, T., Karafiát, M., Burget, L., Černocký, J., & Khudanpur, S. (2010) · Interspeech 2010 · doi:10.21437/interspeech.2010-343 · Cited by 5,390 · see citing works
A Neural Probabilistic Language Model
Bengio, Y., Ducharme, R., Vincent, P., Jauvin, C., Jaz, K., Hofmann, T., Poggio, T., & Shawe-Taylor, J. S. (2000) · Cited by 3,451 · see citing works
Learning to forget: continual prediction with LSTM
Gers, F. A. (1999) · doi:10.1049/cp:19991218 · Cited by 2,436 · see citing works
Extensions of recurrent neural network language model
Mikolov, T., Kombrink, S., Burget, L., Černocký, J., & Khudanpur, S. P. (2011) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2011.5947611 · Cited by 1,594 · see citing works
Improved backing-off for M-gram language modeling
Kneser, R., & Ney, H. (2002) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.1995.479394 · Cited by 1,471 · see citing works
Self-citation(2005) · Cited by 823 · see citing works
Learning Representations by Back-Propagating Errors
Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (2002) · The MIT Press eBooks · doi:10.7551/mitpress/1888.003.0013 · Cited by 794 · see citing works
Learning Recurrent Neural Networks with Hessian-Free Optimization
Martens, J., & Sutskever, I. (2011) · Cited by 525 · see citing works
Continuous space language models
Schwenk, H. (2006) · Computer Speech & Language · doi:10.1016/j.csl.2006.09.003 · Cited by 524 · see citing works
RNNLM - Recurrent Neural Network Language Modeling Toolkit
Mikolov, T., Kombrink, S., Deoras, A., Burget, L., & Černocký, J. (2011) · Cited by 259 · see citing works
Classes for fast maximum entropy training
Goodman, J. T. (2002) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2001.940893 · Cited by 225 · see citing works
Training Continuous Space Language Models: Some Practical Issues
Le, H.-S., Allauzen, A., Wisniewski, G., & Yvon, F. (2010) · Empirical Methods in Natural Language Processing · Cited by 30 · see citing works
Performance analysis of Neural Networks in combination with n-gram language models
Oparin, I., Sundermeyer, M., Ney, H., et al. (2012) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2012.6289044 · Cited by 29 · see citing works
Self-citation
Flagged first, then by citations
Research works printed but not linked: 1
- Gers, F. A., Schraudolph, N. N., Schmidhuber, J., "Learning Pre- cise Timing with LSTM Recurrent Networks", Journal of Ma- chine Learning Research 3, 2002, pp. 115-143
Cited by
Sorted by citations
Fundamentals of Recurrent Neural Network (RNN) and Long Short-Term Memory (LSTM) Network
Sherstinsky, A. (2020) · Physica D: Nonlinear Phenomena · doi:10.1016/j.physd.2019.132306 · Cited by 5,680 · see citing works
DeepLog: Anomaly Detection and Diagnosis from System Logs through Deep Learning
Du, M., Li, F., Zheng, G., & Srikumar, V. (2017) · Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/3133956.3134015 · Cited by 1,835 · see citing works
A Survey on Federated Learning Systems: Vision, Hype and Reality for Data Privacy and Protection
Li, Q., Wen, Z., Wu, Z., Hu, S., Wang, N., Li, Y., Liu, X., & He, B. (2023) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2021.3124599 · Cited by 1,407 · see citing works
The Curious Case of Neural Text Degeneration
Holtzman, A., Buys, J., Du, L., Forbes, M., & Choi, Y. (2019) · arXiv · doi:10.48550/arxiv.1904.09751 · Cited by 1,304 · see citing works
Deep learning in video multi-object tracking: A survey
Ciaparrone, G., Luque Sánchez, F., Tabik, S., Troiano, L., Tagliaferri, R., & Herrera, F. (2020) · Neurocomputing · doi:10.1016/j.neucom.2019.11.023 · Cited by 665 · see citing works
Deep Recurrent Neural Networks for Human Activity Recognition
Murad, A., & Pyun, J.-Y. (2017) · Sensors · doi:10.3390/s17112556 · Cited by 524 · see citing works
The Microsoft 2017 Conversational Speech Recognition System
Xiong, W., Wu, L., Alleva, F., Droppo, J., Huang, X., & Stolcke, A. (2018) · 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2018.8461870 · Cited by 472 · see citing works
Asleep at the Keyboard? Assessing the Security of GitHub Copilot’s Code Contributions
Pearce, H., Ahmad, B., Tan, B., Dolan-Gavitt, B., & Karri, R. (2022) · 2022 IEEE Symposium on Security and Privacy (SP) · doi:10.1109/sp46214.2022.9833571 · Cited by 476 · see citing works
Toward Human Parity in Conversational Speech Recognition
Xiong, W., Droppo, J., Huang, X., Seide, F., Seltzer, M. L., Stolcke, A., Yu, D., & Zweig, G. (2017) · IEEE/ACM Transactions on Audio, Speech, and Language Processing · doi:10.1109/taslp.2017.2756440 · Cited by 245 · see citing works
Incorporating Context into Language Encoding Models for fMRI
Jain, S., & Huth, A. G. (2018) · doi:10.1101/327601 · Cited by 155 · see citing works
Asleep at the keyboard? Assessing the security of GitHub copilot’s code contributions
Pearce, H., Ahmad, B., Tan, B., Dolan-Gavitt, B., & Karri, R. (2025) · Communications of the ACM · doi:10.1145/3610721 · Cited by 140 · see citing works
“DL architectures that demonstrate good fits for NLP include LSTMs [13], RNNs [14], and Transformers [15] that have paved the way for models such as BERT [16], GPT-2 [17], and GPT-3 [18].”
MethodSecuring Industrial Internet of Things Against Botnet Attacks Using Hybrid Deep Learning Approach
Hasan, T., Malik, J., Bibi, I., Khan, W. U., Al-Wesabi, F. N., Dev, K., & Huang, G. (2023) · IEEE Transactions on Network Science and Engineering · doi:10.1109/tnse.2022.3168533 · Cited by 121 · see citing works
Deep Fisher discriminant learning for mobile hand gesture recognition
Li, C., Xie, C., Zhang, B., Chen, C., & Han, J. (2018) · Pattern Recognition · doi:10.1016/j.patcog.2017.12.023 · Cited by 72 · see citing works
Efficient Nearest Neighbor Language Models
He, J., Neubig, G., & Berg-Kirkpatrick, T. (2021) · Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2021.emnlp-main.461 · Cited by 59 · see citing works
“Modern neural language models (NLMs) based on recurrent (Mikolov et al., 2010; Sundermeyer et al., 2012) or selfattentional (Vaswani et al.”
BackgroundInfluentialCharacter-Level Language Modeling with Deeper Self-Attention
Al-Rfou, R., Choe, D., Constant, N., Guo, M., & Jones, L. (2019) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v33i01.33013159 · Cited by 56 · see citing works
Hunting for Insider Threats Using LSTM-Based Anomaly Detection
Villarreal-Vasquez, M., Modelo-Howard, G., Dube, S., & Bhargava, B. (2023) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2021.3135639 · Cited by 52 · see citing works
Latent Relation Language Models
Hayashi, H., Hu, Z., Xiong, C., & Neubig, G. (2020) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v34i05.6298 · Cited by 44 · see citing works
Multi-modality Sensor Data Classification with Selective Attention
Zhang, X., Yao, L., Huang, C., Wang, S., Tan, M., Long, G., & Wang, C. (2018) · Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2018/432 · Cited by 41 · see citing works
Time Series-Based GHG Emissions Prediction for Smart Homes
Riekstin, A. C., Langevin, A., Dandres, T., Gagnon, G., & Cheriet, M. (2020) · IEEE Transactions on Sustainable Computing · doi:10.1109/tsusc.2018.2886164 · Cited by 41 · see citing works
Studying the difference between natural and programming language corpora
Casalnuovo, C., Sagae, K., & Devanbu, P. (2019) · Empirical Software Engineering · doi:10.1007/s10664-018-9669-7 · Cited by 33 · see citing works
Hou, Y., Tamoto, H., & Miyashita, H. (2024) · Extended Abstracts of the CHI Conference on Human Factors in Computing Systems · doi:10.1145/3613905.3650839 · Cited by 31 · see citing works
“While transformers possess excellent self-attention mechanisms, outperforming recurrent neural networks (RNNs) [15] and long short-term memory models (LSTM) [23], they fail to replicate human behavioral dynamics.”
BackgroundVision to Language: Methods, Metrics and Datasets
Sharif, N., Nadeem, U., Shah, S. A. A., Bennamoun, M., & Liu, W. (2020) · Learning and Analytics in Intelligent Systems · doi:10.1007/978-3-030-49724-8_2 · Cited by 9 · see citing works
Pinhanez, C. S., Candello, H., Cavalin, P., Pichiliani, M. C., Appel, A. P., Alves Ribeiro, V. H., Nogima, J., de Bayser, M., Guerra, M., Ferreira, H., et al. (2021) · Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3411764.3445368 · Cited by 7 · see citing works
- Authors
- Sundermeyer, M., Schlüter, R., & Ney, H.
- Affiliations
- RWTH Aachen University (DE)
- Venue
- Interspeech 2012
- Publisher
- ISCA
- Published
- 2012-09-09
- Pages
- 194-197
- ISSN
- 2308-457X (ISSN-L)
- OpenAlex
- W2402268235
- Cited by
- 1,963
- References
- 20
- Citations by year
- 2026: 14 · 2025: 84 · 2024: 115 · 2023: 175 · 2022: 238 · 2021: 276 · 2020: 306 · 2019: 247 · 2018: 210 · 2017: 141 · 2016: 87 · 2015: 42 · 2014: 21 · 2013: 5 · 2012: 1
- FWCI
- 16.93
- Type
- conference paper
- Language
- en
- Open access
- Locations
- Interspeech (USB) (publishedVersion)
- RWTH Publications (RWTH Aachen) (submittedVersion)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Speech Recognition and Synthesis
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
- Keywords
- language modeling, long short-term memory networks, recurrent neural networks, perplexity, word error rate, speech recognition, English language, French language, next-word prediction
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
