GPT or BERT: why not both?

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Charpentier, Lucas Georges Gabriel, Samuel, David
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912171154210816
author Charpentier, Lucas Georges Gabriel
Samuel, David
author_facet Charpentier, Lucas Georges Gabriel
Samuel, David
contents We present a simple way to merge masked language modeling with causal language modeling. This hybrid training objective results in a model that combines the strengths of both modeling paradigms within a single transformer stack: GPT-BERT can be transparently used like any standard causal or masked language model. We test the pretraining process that enables this flexible behavior on the BabyLM Challenge 2024. The results show that the hybrid pretraining outperforms masked-only or causal-only models. We openly release the models, training corpora and code.
format Preprint
id arxiv_https___arxiv_org_abs_2410_24159
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle GPT or BERT: why not both?
Charpentier, Lucas Georges Gabriel
Samuel, David
Computation and Language
We present a simple way to merge masked language modeling with causal language modeling. This hybrid training objective results in a model that combines the strengths of both modeling paradigms within a single transformer stack: GPT-BERT can be transparently used like any standard causal or masked language model. We test the pretraining process that enables this flexible behavior on the BabyLM Challenge 2024. The results show that the hybrid pretraining outperforms masked-only or causal-only models. We openly release the models, training corpora and code.
title GPT or BERT: why not both?
topic Computation and Language
url https://arxiv.org/abs/2410.24159