BriefGPT.xyz
Feb, 2023
使用门控融合方法提高自然语言处理模型升级的预测向后兼容性
Improving Prediction Backward-Compatiblility in NLP Model Upgrade with Gated Fusion
HTML
PDF
Yi-An Lai, Elman Mansimov, Yuqing Xie, Yi Zhang
TL;DR
本文提出了一种名为“Gated Fusion”的新方法，它通过学习在旧模型和新模型之间混合预测来促进向后兼容的模型升级，实验结果表明，该方法平均降低了62%的回归错误，比最强基线平均提高了25%。
Abstract
When upgrading
neural models
to a newer version, new errors that were not encountered in the legacy version can be introduced, known as
regression errors
. This inconsistent behavior during model upgrade often out
→