Universal Language Model Fine-tuning for Text Classification
Open accessThe DOI matches its Crossref record; 56 of the 59 references it cites were found in a registry or an index.
Abstract
Inductive transfer learning has greatly impacted computer vision, but existing approaches in NLP still require task-specific modifications and training from scratch.We propose Universal Language Model Fine-tuning (ULMFiT), an effective transfer learning method that can be applied to any task in NLP, and introduce techniques that are key for fine-tuning a language model.Our method significantly outperforms the state-of-the-art on six text classification tasks, reducing the error by 18-24% on the majority of datasets.Furthermore, with only 100 labeled examples, it matches the performance of training from scratch on 100× more data.We opensource our pretrained models and code 1 .
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
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 cited source says what the claim attributes to it.
In the citing paper
“we can utilize the technique of fine-tuning pre-trained language models32–35 with family-specific sequences to further improve ProGen’s ability to capture the distribution of local sequence neighborhoods corresponding to the protein family.”
From Large language models generate functional protein sequences across diverse families (2023) · cites this paper as [33]
The cited source says what the claim attributes to it.
In the citing paper
“We also experimented with fine-tuning [53] to guide the models' opinion, but the fine-tuned models did not consistently produce the intended opinion.”
From Co-Writing with Opinionated Language Models Affects Users’ Views (2023) · cites this paper as [53]
The cited source says what the claim attributes to it.
In the citing paper
“Techniques like slanted triangular learning rates and gradual unfreezing can be used to avoid catastrophic forgetting during fine-tuning [Howard and Ruder, 2018].”
From Authorship Attribution in the Era of LLMs: Problems, Methodologies, and Challenges (2025) · cites this paper as Howard and Ruder, 2018
The cited source says what the claim attributes to it.
In the citing paper
“We also experimented with using a triangular learning rate (Howard and Ruder, 2018), which produced slightly better results but requires knowing the total number of training steps ahead of time.”
From Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer (2019) · cites this paper as Howard and Ruder, 2018
The cited source says what the claim attributes to it.
In the citing paper
“Task-adaptive pretraining has been shown effective (Howard and Ruder, 2018), but is not typically used with the most recent models.”
From Don’t Stop Pretraining: Adapt Language Models to Domains and Tasks (2020) · cites this paper as Howard and Ruder, 2018
1 more claim about this paper
Create a free account to see every claim other papers make about this one.
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
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · cited 1×
“…CV models (including object detection, classification, and segmentation) are rarely trained from scratch, but instead are fine-tuned from models that have been pretrained on ImageNet, MS-COCO, and other datasets (Sharif Razavian et al., 2014; Long et al., 2015a; He et al., 2016; Huang et al., 2017).”
MethodDensely Connected Convolutional Networks
Huang, G., Liu, Z., Van Der Maaten, L., & Weinberger, K. Q. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.243 · Cited by 48,566
Fully convolutional networks for semantic segmentation
Long, J., Shelhamer, E., & Darrell, T. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298965 · Cited by 38,834
Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
Ioffe, S., & Szegedy, C. (2015) · arXiv · doi:10.48550/arxiv.1502.03167 · Cited by 27,818 · cited 1×
“Following standard practice for CV classifiers, each block uses batch normalization (Ioffe and Szegedy, 2015) and dropout, with ReLU activations for the intermediate layer and a softmax activation that outputs a probability distribution over target classes at the last layer.”
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 2×
“For inductive transfer, fine-tuning pretrained word embeddings (Mikolov et al., 2013), a simple transfer technique that only targets a model’s first layer, has had a large impact in practice and is used in most state-of-the-art models.”
BackgroundMethodPan, S. J., & Yang, Q. (2010) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2009.191 · Cited by 23,387 · cited 1×
“We are interested in the most general inductive transfer learning setting for NLP (Pan and Yang, 2010): Given a static source task T S and any target task T T with T S 6 = T T , we would like to improve performance on T T .”
BackgroundDeep Contextualized Word Representations
Peters, M., Neumann, M., Iyyer, M., Gardner, M., Clark, C., Lee, K., & Zettlemoyer, L. (2018) · Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers) · doi:10.18653/v1/n18-1202 · Cited by 11,983 · cited 9×
“Specifically, Peters et al. (2018) require engineered custom architectures, while we show state-of-the-art performance with the same basic architecture across a range of tasks.”
BackgroundMethodInfluentialFully convolutional networks for semantic segmentation
Shelhamer, E., Long, J., & Darrell, T. (2017) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2016.2572683 · Cited by 11,898
Learning Word Vectors for Sentiment Analysis
Maas, A. L., Daly, R. E., Pham, P. T., Huang, D., Ng, A. Y., & Potts, C. (2011) · Cited by 6,960
SGDR: Stochastic Gradient Descent with Warm Restarts
Loshchilov, I., & Hutter, F. (2016) · arXiv · doi:10.48550/arxiv.1608.03983 · Cited by 5,309 · cited 2×
“5 In Section 5, we compare against aggressive cosine annealing, a similar schedule that has recently been used to achieve state-of-the-art performance in CV (Loshchilov and Hutter, 2017).”
MethodAn overview of gradient descent optimization algorithms
Ruder, S. (2016) · arXiv · doi:10.48550/arxiv.1609.04747 · Cited by 4,962 · cited 1×
“For context, the regular stochastic gradient descent (SGD) update of a model’s parameters θ at time step t looks like the following (Ruder, 2016):”
BackgroundSelf-citationLearning Transferable Features with Deep Adaptation Networks
Long, M., Cao, Y., Wang, J., & Jordan, M. I. (2015) · arXiv · doi:10.48550/arxiv.1502.02791 · Cited by 4,505 · cited 1×
“For this reason, most work in CV focuses on transferring the last layers of the model (Long et al., 2015b).”
BackgroundCNN Features Off-the-Shelf: An Astounding Baseline for Recognition
Razavian, A. S., Azizpour, H., Sullivan, J., & Carlsson, S. (2014) · 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops · doi:10.1109/cvprw.2014.131 · Cited by 4,451 · cited 2×
“…CV models (including object detection, classification, and segmentation) are rarely trained from scratch, but instead are fine-tuned from models that have been pretrained on ImageNet, MS-COCO, and other datasets (Sharif Razavian et al., 2014; Long et al., 2015a; He et al., 2016; Huang et al., 2017).”
MethodDeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition
Donahue, J., Jia, Y., Vinyals, O., Hoffman, J., Zhang, N., Tzeng, E., & Darrell, T. (2013) · arXiv · doi:10.48550/arxiv.1310.1531 · Cited by 4,019 · cited 5×
“approach has been superseded by fine-tuning either the last (Donahue et al., 2014) or several of the last layers of a pretrained model and leaving the remaining layers frozen (Long et al.”
MethodInfluentialCharacter-level Convolutional Networks for Text Classification
Zhang, X., Zhao, J., & LeCun, Y. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1509.01626 · Cited by 3,205 · cited 2×
“Topic classification For topic classification, we evaluate on the large-scale AG news and DBpedia ontology datasets created by Zhang et al. (2015).”
MethodCyclical Learning Rates for Training Neural Networks
Smith, L. N. (2017) · 2017 IEEE Winter Conference on Applications of Computer Vision (WACV) · doi:10.1109/wacv.2017.58 · Cited by 2,670 · cited 1×
“STLR modifies triangular learning rates (Smith, 2017) with a short increase and a long decay period, which we found key for good performance.”
BackgroundWhy Does Unsupervised Pre-training Help Deep Learning?
Erhan, D., Bengio, Y., Courville, A. C., Manzagol, P.-A., Vincent, P., & Bengio, S. (2010) · Journal of Machine Learning Research · Cited by 2,631
Salakhutdinov, R., & Hinton, G. E. (2009) · International Conference on Artificial Intelligence and Statistics · Cited by 2,605
Fully Convolutional Networks for Semantic Segmentation
Long, J., Shelhamer, E., & Darrell, T. J. (2014) · arXiv (Cornell University) · doi:10.48550/arxiv.1411.4038 · Cited by 2,489
Supervised Learning of Universal Sentence Representations from Natural Language Inference Data
Conneau, A., Kiela, D., Schwenk, H., Barrault, L., & Bordes, A. (2017) · Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d17-1070 · Cited by 2,049 · cited 5×
“…we believe that language model fine-tuning will be particularly useful in the following settings compared to existing transfer learning approaches (Conneau et al., 2017; Mc-Cann et al., 2017; Peters et al., 2018): a) NLP for non-English languages, where training data for supervised pretraining…”
BackgroundMethodInfluential(2007) · Cited by 2,018
Hypercolumns for object segmentation and fine-grained localization
Hariharan, B., Arbelaez, P., Girshick, R., & Malik, J. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298642 · Cited by 1,511 · cited 1×
“This method is known as hypercolumns (Hariharan et al., 2015) in CV and is used by Peters et al. (2017), Peters et al. (2018), Wieting and Gimpel (2017), Conneau et al. (2017), and McCann et al. (2017) who use language modeling, paraphrasing, entailment, and Machine Translation (MT) respectively…”
MethodNgai, E. W. T., Hu, Y., Wong, Y. H., Chen, Y., & Sun, X. (2011) · Decision Support Systems · doi:10.1016/j.dss.2010.08.006 · Cited by 1,361 · cited 1×
“…is a category of Natural Language Processing (NLP) tasks with real-world applications such as spam, fraud, and bot detection (Jindal and Liu, 2007; Ngai et al., 2011; Chu et al., 2012), emergency response (Caragea et al., 2011), and commercial document classification, such as for legal discovery…”
BackgroundExploring the Limits of Weakly Supervised Pretraining
Mahajan, D., Girshick, R., Ramanathan, V., He, K., Paluri, M., Li, Y., Bharambe, A., & van der Maaten, L. (2018) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-01216-8_12 · Cited by 1,224
Semi-supervised Sequence Learning
Dai, A. M., & Le, Q. V. (2015) · arXiv · doi:10.48550/arxiv.1511.01432 · Cited by 994 · cited 3×
“Dai and Le (2015) first proposed fine-tuning a language model (LM) but require millions of in-domain documents to achieve good performance, which severely limits its applicability.”
BackgroundMethodInfluentialDeep Pyramid Convolutional Neural Networks for Text Categorization
Johnson, R., & Zhang, T. (2017) · Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/p17-1052 · Cited by 823 · cited 2×
“Pre-processing We use the same pre-processing as in earlier work (Johnson and Zhang, 2017; McCann et al., 2017).”
MethodEstimation of Dependences Based on Empirical Data.
Shoesmith, E., Vapnik, V., & Kotz, S. (1984) · The Statistician · doi:10.2307/2988246 · Cited by 764
Assessing the Ability of LSTMs to Learn Syntax-Sensitive Dependencies
Linzen, T., Dupoux, E., & Goldberg, Y. (2016) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00115 · Cited by 737 · cited 4×
“syntax-sensitive dependencies (Linzen et al., 2016) to create a model”
BackgroundInfluentialA Model of Inductive Bias Learning
Baxter, J. (2000) · Cited by 730
Felbo, B., Mislove, A. E., Søgaard, A., Rahwan, I., & Lehmann, S. (2017) · Research at the University of Copenhagen (University of Copenhagen) · Cited by 712
Regularizing and Optimizing LSTM Language Models
Merity, S., Keskar, N. S., & Socher, R. (2017) · arXiv · doi:10.48550/arxiv.1708.02182 · Cited by 697 · cited 9×
“We use the AWD-LSTM language model (Merity et al., 2017a) with an embedding size of 400 , 3 layers, 1150 hidden activations per layer, and a BPTT batch size of 70 .”
MethodInfluentialDeep Biaffine Attention for Neural Dependency Parsing
Atmani, M., & Lafourcade, M. (2016) · arXiv (Cornell University) · doi:10.4230/oasics.ldk.2021.24 · Cited by 693 · cited 1×
“99 , similar to (Dozat and Manning, 2017).”
Semi-supervised sequence tagging with bidirectional language models
Peters, M., Ammar, W., Bhagavatula, C., & Power, R. (2017) · Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/p17-1161 · Cited by 655 · cited 4×
“Recent approaches that concatenate embeddings derived from other tasks with the input at different layers (Peters et al., 2017; McCann et al., 2017; Peters et al., 2018) still train the main task model from scratch and treat pretrained embeddings as fixed parameters, limiting their usefulness.”
BackgroundMethodInfluentialDetecting Automation of Twitter Accounts: Are You a Human, Bot, or Cyborg?
Chu, Z., Gianvecchio, S., Wang, H., & Jajodia, S. (2012) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2012.75 · Cited by 634 · cited 2×
“Text classification is a category of Natural Language Processing (NLP) tasks with real-world applications such as spam, fraud, and bot detection (Jindal and Liu, 2007; Ngai et al., 2011; Chu et al., 2012), emergency response (Caragea et al.”
BackgroundA model of inductive bias learning
Baxter, J. (2000) · Journal of Artificial Intelligence Research · doi:10.1613/jair.731 · Cited by 582 · cited 1×
“Formally, language modeling induces a hypothesis space H that should be useful for many other NLP tasks (Vapnik and Kotz, 1982; Baxter, 2000).”
BackgroundLearned in Translation: Contextualized Word Vectors
McCann, B., Bradbury, J., Xiong, C., & Socher, R. (2017) · arXiv · doi:10.48550/arxiv.1708.00107 · Cited by 572 · cited 16×
“On IMDb, we reduce the error dramatically by 43.9% and 22% with regard to CoVe and the state-of-the-art respectively.”
BackgroundMethodInfluentialDeep Biaffine Attention for Neural Dependency ParsingPreprint
Dozat, T., & Manning, C. D. (2016) · arXiv (Cornell University) · Cited by 515
Colorless Green Recurrent Networks Dream Hierarchically
Gulordava, K., Bojanowski, P., Grave, E., Linzen, T., & Baroni, M. (2018) · Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers) · doi:10.18653/v1/n18-1108 · Cited by 465 · cited 2×
“, 2016), hierarchical relations (Gulordava et al., 2018), and sentiment (Radford et al.”
BackgroundJindal, N., & Liu, B. (2007) · Proceedings of the 16th international conference on World Wide Web · doi:10.1145/1242572.1242759 · Cited by 449 · cited 1×
“Text classification is a category of Natural Language Processing (NLP) tasks with real-world applications such as spam, fraud, and bot detection (Jindal and Liu, 2007; Ngai et al., 2011; Chu et al., 2012), emergency response (Caragea et al.”
BackgroundGradNorm: Gradient Normalization for Adaptive Loss Balancing in Deep Multitask Networks
Chen, Z., Badrinarayanan, V., Lee, C., & Rabinovich, A. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1711.02257 · Cited by 443 · cited 1×
“MTL requires the tasks to be trained from scratch every time, which makes it inefficient and often requires careful weighting of the task-specific objective functions (Chen et al., 2017).”
Text Classification Improved by Integrating Bidirectional LSTM with Two-dimensional Max Pooling
Zhou, P., Qi, Z., Zheng, S., Xu, J., Bao, H., & Xu, B. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1611.06639 · Cited by 434 · cited 1×
“(Johnson and Zhang, 2016) 5.9 TBCNN (Mou et al., 2015) 4.0 Virtual (Miyato et al., 2016) 5.9 LSTM-CNN (Zhou et al., 2016) Johnson and Zhang (2017).”
InfluentialPointer Sentinel Mixture Models
Merity, S., Xiong, C., Bradbury, J., & Socher, R. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1609.07843 · Cited by 427 · cited 9×
“We use the AWD-LSTM language model (Merity et al., 2017a) with an embedding size of 400 , 3 layers, 1150 hidden activations per layer, and a BPTT batch size of 70 .”
MethodInfluentialEmpower Sequence Labeling with Task-Aware Neural Language Model
Liu, L., Shang, J., Ren, X., Xu, F., Gui, H., Peng, J., & Han, J. (2018) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v32i1.12006 · Cited by 333
Learning to Generate Reviews and Discovering Sentiment
Radford, A., Jozefowicz, R., & Sutskever, I. (2017) · arXiv · doi:10.48550/arxiv.1704.01444 · Cited by 330 · cited 1×
“…modeling can be seen as the ideal source task and a counter-part of ImageNet for NLP: It captures many facets of language relevant for downstream tasks, such as long-term dependencies (Linzen et al., 2016), hierarchical relations (Gulordava et al., 2018), and sentiment (Radford et al., 2017).”
BackgroundSupervised and Semi-Supervised Text Categorization using LSTM for Region EmbeddingsPreprint
Johnson, R., & Zhang, T. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1602.02373 · Cited by 288 · cited 2×
“This is promising as the existing state-of-the-art requires complex architectures (Peters et al., 2018), multiple forms of attention (McCann et al., 2017) and sophisticated embedding schemes (Johnson and Zhang, 2016), while our method employs a regular LSTM with dropout.”
MethodInfluentialWhat makes ImageNet good for transfer learning?
Huh, M., Agrawal, P., & Efros, A. A. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1608.08614 · Cited by 266 · cited 1×
“…model pretraining and fine-tuning and make them more scalable: for ImageNet, predicting far fewer classes only incurs a small performance drop (Huh et al., 2016), while recent work shows that an alignment be-tween source and target task label sets is important (Mahajan et al., 2018)—focusing…”
BackgroundSemi-supervised Multitask Learning for Sequence Labeling
Rei, M. (2017) · Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/p17-1194 · Cited by 257 · cited 1×
“This is the approach taken by Rei (2017) and Liu et al. (2018) who add a language modeling objective to the model that is trained jointly with the main task model.”
MethodUNITN: Training Deep Convolutional Neural Network for Twitter Sentiment Classification
Severyn, A., & Moschitti, A. (2015) · Proceedings of the 9th International Workshop on Semantic Evaluation (SemEval 2015) · doi:10.18653/v1/s15-2079 · Cited by 201 · cited 1×
“Fine-tuning Fine-tuning has been used successfully to transfer between similar tasks, e.g. in QA (Min et al., 2017), for distantly supervised sentiment analysis (Severyn and Moschitti, 2015), or MT domains (Sennrich et al., 2015) but has been shown to fail between unrelated ones (Mou et al., 2016).”
BackgroundAdversarial Training Methods for Semi-Supervised Text Classification
Miyato, T., Dai, A. M., & Goodfellow, I. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1605.07725 · Cited by 179 · cited 1×
“(Johnson and Zhang, 2016) 5.9 TBCNN (Mou et al., 2015) 4.0 Virtual (Miyato et al., 2016) 5.9 LSTM-CNN (Zhou et al., 2016) Johnson and Zhang (2017).”
InfluentialClassifying text messages for the haiti earthquake
Caragea, C., McNeese, N. J., Jaiswal, A., Traylor, G., Kim, H., Mitra, P., Wu, D., Tapia, A., Giles, C. L., Jansen, B. J. J., et al. (2011) · Cited by 172
Question Answering through Transfer Learning from Large Fine-grained Supervision Data
Min, S., Seo, M., & Hajishirzi, H. (2017) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/p17-2081 · Cited by 157 · cited 1×
“Fine-tuning Fine-tuning has been used successfully to transfer between similar tasks, e.g. in QA (Min et al., 2017), for distantly supervised sentiment analysis (Severyn and Moschitti, 2015), or MT domains (Sennrich et al., 2015) but has been shown to fail between unrelated ones (Mou et al., 2016).”
BackgroundDiscriminative Neural Sentence Modeling by Tree-Based Convolution
Mou, L., Peng, H., Li, G., Xu, Y., Zhang, L., & Jin, Z. G. (2015) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/d15-1279 · Cited by 144 · cited 1×
“(Johnson and Zhang, 2016) 5.9 TBCNN (Mou et al., 2015) 4.0 Virtual (Miyato et al., 2016) 5.9 LSTM-CNN (Zhou et al., 2016) Johnson and Zhang (2017).”
InfluentialRevisiting Recurrent Networks for Paraphrastic Sentence Embeddings
Wieting, J., & Gimpel, K. (2017) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/p17-1190 · Cited by 89 · cited 1×
“This method is known as hypercolumns (Hariharan et al., 2015) in CV and is used by Peters et al. (2017), Peters et al. (2018), Wieting and Gimpel (2017), Conneau et al. (2017), and McCann et al. (2017) who use language modeling, paraphrasing, entailment, and Machine Translation (MT) respectively…”
MethodDocument categorization in legal electronic discovery: computer classification vs. manual review
Roitblat, H. L., Kershaw, A., & Oot, P. (2009) · Journal of the American Society for Information Science and Technology · doi:10.1002/asi.21233 · Cited by 64 · cited 1×
“…Language Processing (NLP) tasks with real-world applications such as spam, fraud, and bot detection (Jindal and Liu, 2007; Ngai et al., 2011; Chu et al., 2012), emergency response (Caragea et al., 2011), and commercial document classification, such as for legal discovery (Roitblat et al., 2010).”
BackgroundAssessing the Ability of LSTMs to Learn Syntax-Sensitive DependenciesPreprint
Linzen, T., Dupoux, E., & Goldberg, Y. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1611.01368 · Cited by 25 · cited 4×
“syntax-sensitive dependencies (Linzen et al., 2016) to create a model”
BackgroundInfluential(2011) · IEEE International Reliability Physics Symposium proceedings · doi:10.1109/irps.2011.5784441 · Cited by 7
Research works printed but not linked: 6
- [19]19. Nitin Jindal and Bing Liu. 2007. Review spam de- tection. In Proceedings of the 16th international conference on World Wide Web. ACM, pages 1189- 1190.
- [35]35. Lili Mou, Zhao Meng, Rui Yan, Ge Li, Yan Xu, Lu Zhang, and Zhi Jin. 2016. How Transferable are Neural Networks in NLP Applications? Proceed- ings of 2016 Conference on Empirical Methods in Natural Language Processing .
- [40]40. Matthew E Peters, Mark Neumann, Mohit Iyyer, Matt Gardner, Christopher Clark, Kenton Lee, and Luke Zettlemoyer. 2018. Deep contextualized word rep- resentations. In Proceedings of NAACL 2018.
- [46]46. Rico Sennrich, Barry Haddow, and Alexandra Birch. 2015. Improving neural machine translation models with monolingual data. arXiv preprint arXiv:1511.06709 . · arXiv
- [52]52. John Wieting and Kevin Gimpel. 2017. Revisiting Re- current Networks for Paraphrastic Sentence Embed- dings. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (ACL 2017).
- [53]53. Jason Yosinski, Jeff Clune, Yoshua Bengio, and Hod Lipson. 2014. How transferable are features in deep neural networks? In Advances in neural information processing systems. pages 3320-3328.
Cited by
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2018) · arXiv · doi:10.48550/arxiv.1810.04805 · Cited by 73,173
“A recent trend in transfer learning from language models (LMs) is to pre-train some model architecture on a LM objective before fine-tuning that same model for a supervised downstream task (Dai and Le, 2015; Howard and Ruder, 2018; Radford et al., 2018).”
BackgroundMethodBERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019) · 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) · doi:10.18653/v1/n19-1423 · Cited by 54,325
“A recent trend in transfer learning from language models (LMs) is to pre-train some model architecture on a LM objective before fine-tuning that same model for a supervised downstream task (Dai and Le, 2015; Howard and Ruder, 2018; Radford et al., 2018).”
BackgroundMethodLanguage Models are Few-Shot Learners
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.14165 · Cited by 23,372
RoBERTa: A Robustly Optimized BERT Pretraining Approach
Liu, Y., Ott, M., Goyal, N., Du, J., Joshi, M., Chen, D., Levy, O., Lewis, M., Zettlemoyer, L., & Stoyanov, V. (2019) · arXiv · doi:10.48550/arxiv.1907.11692 · Cited by 16,515
“Pretraining methods have been designed with different training objectives, including language modeling (Dai and Le, 2015; Peters et al., 2018; Howard and Ruder, 2018), machine translation (McCann et al., 2017), and masked language modeling (Devlin et al., 2019; Lample and Conneau, 2019).”
MethodLearning Transferable Visual Models From Natural Language Supervision
Radford, A., Kim, J. W., Hallacy, C., Ramesh, A., Goh, G., Agarwal, S., Sastry, G., Askell, A., Mishkin, P., Clark, J., et al. (2021) · arXiv · doi:10.48550/arxiv.2103.00020 · Cited by 15,293
“Pre-training methods which learn directly from raw text have revolutionized NLP over the last few years (Dai & Le, 2015; Peters et al., 2018; Howard & Ruder, 2018; Radford et al., 2018; Devlin et al., 2018; Raffel et al., 2019).”
Method
Metadata
- Authors
- Howard, J., & Ruder, S.
- Affiliations
- North Carolina State University (US)
- Ollscoil na Gaillimhe – University of Galway (IE)
- Venue
- Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
- Published
- 2018
- Pages
- 328-339
- ISSN
- 0736-587X (ISSN-L)
- OpenAlex
- W2798812533
- Cited by
- 4,100
- References
- 59
- Citations by year
- 2026: 263 · 2025: 495 · 2024: 451 · 2023: 474 · 2022: 522 · 2021: 721 · 2020: 652 · 2019: 477 · 2018: 43
- FWCI
- 188.51
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- Locations
- Proceedings of the conference - Association for Computational Linguistics. Meeting (publishedVersion, cc-by)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- Research Councils UK (AH/S01179X/1)
- Science Foundation Ireland (SFI/12/RC/, SFI/12/RC/2289., 12/RC/, Grant Number SFI/12/RC/2289, 12/RC/2289)
- Irish Research Council
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
- Keywords
- ULMFiT, transfer learning, text classification, language model fine-tuning, pre-trained language models, inductive transfer learning, low-resource training
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
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.