Skip to yearly menu bar Skip to main content


Unbounded cache model for online language modeling with open vocabulary

Edouard Grave · Moustapha Cisse · Armand Joulin

Pacific Ballroom #83

Keywords: [ Recurrent Networks ] [ Natural Language Processing ]


Recently, continuous cache models were proposed as extensions to recurrent neural network language models, to adapt their predictions to local changes in the data distribution. These models only capture the local context, of up to a few thousands tokens. In this paper, we propose an extension of continuous cache models, which can scale to larger contexts. In particular, we use a large scale non-parametric memory component that stores all the hidden activations seen in the past. We leverage recent advances in approximate nearest neighbor search and quantization algorithms to store millions of representations while searching them efficiently. We conduct extensive experiments showing that our approach significantly improves the perplexity of pre-trained language models on new distributions, and can scale efficiently to much larger contexts than previously proposed local cache models.

Live content is unavailable. Log in and register to view live content