# Language modeling

tags
NLP

## LM with RNNs

Different models have been studied, starting from the initial Recurrent neural network based language model (Mikolov et al. 2011).

LSTM were then used with more success than previous models (Zaremba, Sutskever, and Vinyals 2015).

Recently, transformers seem to have dominated language modeling. However it is not clear if this is due to their real superiority over RNNs or their practical scalability (Merity 2019).

## Text generation

Language models can be used to generate text from a prompt or starting sentence. This is the kind of examples that made models like GPT-2 and GPT-3 famous, because of their ability to generate long sequences of apparently coherent text (Radford et al. 2019; Brown et al. 2020).

## Bibliography

1. . . "Language Models Are Few-Shot Learners". arXiv:2005.14165 [Cs]. http://arxiv.org/abs/2005.14165.

2. . n.d. “Reinforcement Learning as One Big Sequence Modeling Problem”, 15.

3. . . “Recurrent Neural Network Based Language Model”, 4.

4. . . “Generative Language Modeling for Automated Theorem Proving”. arXiv:2009.03393 [Cs, Stat]. http://arxiv.org/abs/2009.03393.

5. . . “Language Models Are Unsupervised Multitask Learners”. OpenAI Blog 1 (8):9.

6. . . “Recurrent Neural Network Regularization”. arXiv:1409.2329 [Cs]. http://arxiv.org/abs/1409.2329.