Gaussian Stochastic Weight Averaging for Bayesian Low-Rank Adaptation of Large Language Models

Fine-tuned LLMs often suffer from overconfidence and poor calibration, particularly when fine-tuned on small datasets. To address these challenges, we propose a simple combination of Low-Rank Adaptation (LoRA) with Gaussian Stochastic Weight Averaging (SWAG), facilitating approximate Bayesian inference in LLMs. Through extensive testing across several NLP benchmarks, we demonstrate that our straightforward and computationally efficient approach improves model generalization and calibration. We further show that our method exhibits greater robustness against distribution shift, as reflected in its performance on out-of-distribution tasks.