Search Results for author: Ernest Pusateri

Found 5 papers, 0 papers with code

Error-driven Pruning of Language Models for Virtual Assistants

no code implementations14 Feb 2021 Sashank Gondala, Lyan Verwimp, Ernest Pusateri, Manos Tsagkias, Christophe Van Gysel

We customize entropy pruning by allowing for a keep list of infrequent n-grams that require a more relaxed pruning threshold, and propose three methods to construct the keep list.

Connecting and Comparing Language Model Interpolation Techniques

no code implementations26 Aug 2019 Ernest Pusateri, Christophe Van Gysel, Rami Botros, Sameer Badaskar, Mirko Hannemann, Youssef Oualil, Ilya Oparin

In this work, we uncover a theoretical connection between two language model interpolation techniques, count merging and Bayesian interpolation.

Language Modelling

Cannot find the paper you are looking for? You can Submit a new open access paper.