Towards Zero-shot Language Modeling

TL;DR

This paper introduces a zero-shot language modeling approach that uses a typologically-informed prior over neural weights, enabling better adaptation to unseen languages, especially with limited data.

Contribution

It proposes a method to construct an informative prior from diverse languages to improve zero-shot and few-shot language modeling performance.

Findings

Prior improves zero-shot and few-shot language modeling

Typological features aid few-shot adaptation

Universal phonological knowledge is captured in the prior

Abstract

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.