Exploring Mode Connectivity for Pre-trained Language Models
This work addresses the problem of understanding the inner workings of PLM adaptation for NLP researchers, but it is incremental as it builds on existing mode connectivity concepts without introducing new methods.
The paper investigates the geometric connections between different minima in pre-trained language models (PLMs) using mode connectivity, analyzing how hyperparameters, tuning methods, and training data affect these connections, and finds that exploring mode connectivity helps understand PLM adaptation dynamics.
Recent years have witnessed the prevalent application of pre-trained language models (PLMs) in NLP. From the perspective of parameter space, PLMs provide generic initialization, starting from which high-performance minima could be found. Although plenty of works have studied how to effectively and efficiently adapt PLMs to high-performance minima, little is known about the connection of various minima reached under different adaptation configurations. In this paper, we investigate the geometric connections of different minima through the lens of mode connectivity, which measures whether two minima can be connected with a low-loss path. We conduct empirical analyses to investigate three questions: (1) how could hyperparameters, specific tuning methods, and training data affect PLM's mode connectivity? (2) How does mode connectivity change during pre-training? (3) How does the PLM's task knowledge change along the path connecting two minima? In general, exploring the mode connectivity of PLMs conduces to understanding the geometric connection of different minima, which may help us fathom the inner workings of PLM downstream adaptation.