# Towards Zero-shot Language Modeling

arXiv: https://arxiv.org/abs/2108.03334 · HTML: https://mlsanity.com/paper/2108.03334/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 27 of its 59 references that are in the map.
- First posted on arXiv 6 Aug 2021.

## Builds on (most cited first, 27 of 27)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [Sequence to Sequence Learning with Neural Networks](https://mlsanity.com/paper/1409.3215/) (2014)
- [Matching Networks for One Shot Learning](https://mlsanity.com/paper/1606.04080/) (2016)
- [Neural Network Acceptability Judgments](https://mlsanity.com/paper/1805.12471/) (2018)
- [An Overview of Multi-Task Learning in Deep Neural Networks](https://mlsanity.com/paper/1706.05098/) (2017)
- [Generating Sequences With Recurrent Neural Networks](https://mlsanity.com/paper/1308.0850/) (2013)
- [Assessing the Ability of LSTMs to Learn Syntax-Sensitive Dependencies](https://mlsanity.com/paper/1611.01368/) (2016)
- [Targeted Syntactic Evaluation of Language Models](https://mlsanity.com/paper/1808.09031/) (2018)
- [Regularizing and Optimizing LSTM Language Models](https://mlsanity.com/paper/1708.02182/) (2017)
- [Online Structured Laplace Approximations For Overcoming Catastrophic Forgetting](https://mlsanity.com/paper/1805.07810/) (2018)
- [To Tune or Not to Tune? Adapting Pretrained Representations to Diverse Tasks](https://mlsanity.com/paper/1903.05987/) (2019)
- [A Theoretically Grounded Application of Dropout in Recurrent Neural Networks](https://mlsanity.com/paper/1512.05287/) (2015)
- [Revisiting Natural Gradient for Deep Networks](https://mlsanity.com/paper/1301.3584/) (2013)
- [Modeling Language Variation and Universals: A Survey on Typological Linguistics for Natural Language Processing](https://mlsanity.com/paper/1807.00914/) (2018)
- [Are All Languages Equally Hard to Language-Model?](https://mlsanity.com/paper/1806.03743/) (2018)
- [What Kind of Language Is Hard to Language-Model?](https://mlsanity.com/paper/1906.04726/) (2019)
- [Contextual Parameter Generation for Universal Neural Machine Translation](https://mlsanity.com/paper/1808.08493/) (2018)
- [Unbiasing Truncated Backpropagation Through Time](https://mlsanity.com/paper/1705.08209/) (2017)
- [Studying the Inductive Biases of RNNs with Synthetic Variations of Natural Languages](https://mlsanity.com/paper/1903.06400/) (2019)
- [An Analysis of Neural Language Modeling at Multiple Scales](https://mlsanity.com/paper/1803.08240/) (2018)
- [Continuous multilinguality with language vectors](https://mlsanity.com/paper/1612.07486/) (2016)
- [Can LSTM Learn to Capture Agreement? The Case of Basque](https://mlsanity.com/paper/1809.04022/) (2018)
- [Metalearning with Hebbian Fast Weights](https://mlsanity.com/paper/1807.05076/) (2018)
- [Bayesian Incremental Learning for Deep Neural Networks](https://mlsanity.com/paper/1802.07329/) (2018)
- [Laplace's Method Approximations for Probabilistic Inference in Belief Networks with Continuous Variables](https://mlsanity.com/paper/1302.6782/) (2013)
- [Polyglot Neural Language Models: A Case Study in Cross-Lingual Phonetic Representation Learning](https://mlsanity.com/paper/1605.03832/) (2016)

## Built on by (most cited first, 3 of 3)

- [Elastic Weight Removal for Faithful and Abstractive Dialogue Generation](https://mlsanity.com/paper/2303.17574/) (2023)
- [Graphemic Normalization of the Perso-Arabic Script](https://mlsanity.com/paper/2210.12273/) (2022)
- [On the Element-Wise Representation and Reasoning in Zero-Shot Image Recognition: A Systematic Survey](https://mlsanity.com/paper/2408.04879/) (2024)

## Abstract (arXiv)

Can we construct a neural model that is inductively biased towards learning human languages? Motivated by this question, we aim at constructing an informative prior over neural weights, in order to adapt quickly to held-out languages in the task of character-level language modeling. We infer this distribution from a sample of typologically diverse training languages via Laplace approximation. The use of such a prior outperforms baseline models with an uninformative prior (so-called "fine-tuning") in both zero-shot and few-shot settings. This shows that the prior is imbued with universal phonological knowledge. Moreover, we harness additional language-specific side information as distant supervision for held-out languages. Specifically, we condition language models on features from typological databases, by concatenating them to hidden states or generating weights with hyper-networks. These features appear beneficial in the few-shot setting, but not in the zero-shot setting. Since the paucity of digital texts affects the majority of the world's languages, we hope that these findings will help broaden the scope of applications for language technology.