Probing Out-of-Distribution Robustness of Language Models with Parameter-Efficient Transfer Learning

TL;DR

This paper investigates how different parameter-efficient transfer learning methods affect the ability of large language models to detect out-of-distribution inputs across various tasks and model sizes.

Contribution

It provides a systematic evaluation of OOD detection capabilities of multiple PETL techniques on different language model scales and tasks.

Findings

Larger models generally improve OOD detection.

Different PETL methods vary in effectiveness for OOD detection.

Model size and transfer method influence robustness to distribution shifts.

Abstract

As the size of the pre-trained language model (PLM) continues to increase, numerous parameter-efficient transfer learning methods have been proposed recently to compensate for the tremendous cost of fine-tuning. Despite the impressive results achieved by large pre-trained language models (PLMs) and various parameter-efficient transfer learning (PETL) methods on sundry benchmarks, it remains unclear if they can handle inputs that have been distributionally shifted effectively. In this study, we systematically explore how the ability to detect out-of-distribution (OOD) changes as the size of the PLM grows or the transfer methods are altered. Specifically, we evaluated various PETL techniques, including fine-tuning, Adapter, LoRA, and prefix-tuning, on three different intention classification tasks, each utilizing various language models with different scales.