Low-rank adaptation (LoRA) has emerged as a new paradigm for cost-efficient fine-tuning of large language models (LLMs). However, fine-tuned LLMs often become overconfident especially when fine-tuned on small datasets. Bayesian methods, with their inherent ability to estimate uncertainty, serve as potent tools to mitigate overconfidence and enhance calibration. In this work, we introduce Laplace-LoRA, which applies a Bayesian approach to the LoRA parameters. Specifically, Laplace-LoRA applies a Laplace approximation to the posterior over the LoRA parameters, considerably improving the calibration of fine-tuned LLMs.
@article{arxiv.2308.13111,
title = {Bayesian Low-rank Adaptation for Large Language Models},
author = {Adam X. Yang and Maxime Robeyns and Xi Wang and Laurence Aitchison},
journal= {arXiv preprint arXiv:2308.13111},
year = {2024}
}