@inproceedings{marszalek-etal-2025-minimal,
    title = "Minimal Ranks, Maximum Confidence: Parameter-efficient Uncertainty Quantification for {L}o{RA}",
    author = "Marsza{\l}ek, Patryk  and
      Ba{\l}azy, Klaudia  and
      Tabor, Jacek  and
      Ku{\'s}mierczyk, Tomasz",
    editor = "Christodoulopoulos, Christos  and
      Chakraborty, Tanmoy  and
      Rose, Carolyn  and
      Peng, Violet",
    booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2025",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/author-page-yu-wang-polytechnic/2025.findings-emnlp.66/",
    doi = "10.18653/v1/2025.findings-emnlp.66",
    pages = "1260--1271",
    ISBN = "979-8-89176-335-7",
    abstract = "Low-Rank Adaptation (LoRA) enables parameter-efficient fine-tuning of large language models by decomposing weight updates into low-rank matrices, significantly reducing storage and computational overhead. While effective, standard LoRA lacks mechanisms for uncertainty quantification, leading to overconfident and poorly calibrated models. Bayesian variants of LoRA address this limitation, but at the cost of a significantly increased number of trainable parameters, partially offsetting the original efficiency gains. Additionally, these models are harder to train and may suffer from unstable convergence.In this work, we propose a novel parameter-efficient Bayesian LoRA via subspace inference, demonstrating that effective uncertainty quantification can be achieved in very low-dimensional parameter spaces. The proposed method achieves strong performance with improved calibration and generalization while maintaining computational efficiency. Our empirical findings show that, with the appropriate projection of the weight space: (1) uncertainty can be effectively modeled in a low-dimensional space, and (2) weight covariances exhibit low ranks."
}