Bayesian low-rank adaptation for large language models

Published: 23 Oct 2023, Last Modified: 28 Nov 2023SoLaR SpotlightEveryoneRevisionsBibTeX
Keywords: Large language models, Bayesian deep learning, uncertainty calibration
Abstract: 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.
Submission Number: 53