Bayesian Low-rank Adaptation for Large Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yang, Adam X., Robeyns, Maxime, Wang, Xi, Aitchison, Laurence
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914667379556352
author Yang, Adam X.
Robeyns, Maxime
Wang, Xi
Aitchison, Laurence
author_facet Yang, Adam X.
Robeyns, Maxime
Wang, Xi
Aitchison, Laurence
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2308_13111
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Bayesian Low-rank Adaptation for Large Language Models
Yang, Adam X.
Robeyns, Maxime
Wang, Xi
Aitchison, Laurence
Machine Learning
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.
title Bayesian Low-rank Adaptation for Large Language Models
topic Machine Learning
url https://arxiv.org/abs/2308.13111