Subword Pooling Makes a Difference

by   Judit Acs, et al.

Contextual word-representations became a standard in modern natural language processing systems. These models use subword tokenization to handle large vocabularies and unknown words. Word-level usage of such systems requires a way of pooling multiple subwords that correspond to a single word. In this paper we investigate how the choice of subword pooling affects the downstream performance on three tasks: morphological probing, POS tagging and NER, in 9 typologically diverse languages. We compare these in two massively multilingual models, mBERT and XLM-RoBERTa. For morphological tasks, the widely used `choose the first subword' is the worst strategy and the best results are obtained by using attention over the subwords. For POS tagging both of these strategies perform poorly and the best choice is to use a small LSTM over the subwords. The same strategy works best for NER and we show that mBERT is better than XLM-RoBERTa in all 9 languages. We publicly release all code, data and the full result tables at <>.



There are no comments yet.


page 7

page 8


BNLP: Natural language processing toolkit for Bengali language

BNLP is an open source language processing toolkit for Bengali language ...

Evaluating Contextualized Language Models for Hungarian

We present an extended comparison of contextualized language models for ...

LINSPECTOR: Multilingual Probing Tasks for Word Representations

Despite an ever growing number of word representation models introduced ...

Trankit: A Light-Weight Transformer-based Toolkit for Multilingual Natural Language Processing

We introduce Trankit, a light-weight Transformer-based Toolkit for multi...

Enhancing deep neural networks with morphological information

Currently, deep learning approaches are superior in natural language pro...

Sequence Labeling: A Practical Approach

We take a practical approach to solving sequence labeling problem assumi...

Reporting Score Distributions Makes a Difference: Performance Study of LSTM-networks for Sequence Tagging

In this paper we show that reporting a single performance score is insuf...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.