Kim, Kwonyoung Park, Jungin Kim, Jin Kwon, Hyeongjun Sohn, Kwanghoon 2026-02-25T00:30:17Z 2026-02-25T00:30:17Z 2026-02-20 2025 1063-6919 https://pubs.kist.re.kr/handle/201004/154353 Parameter-efficient tuning (PET) aims to transfer pre-trained foundation models to downstream tasks by learning a small number of parameters. Compared to traditional fine-tuning, which updates the entire model, PET significantly reduces storage and transfer costs for each task regardless of exponentially increasing pre-trained model capacity. However, most PET methods inherit the inference latency of their large backbone models and often introduce additional computational overhead due to additional modules (e.g. adapters), limiting their practicality for compute-intensive applications. In this paper, we propose Faster Parameter-Efficient Tuning (FPET), a novel approach that enhances inference speed and training efficiency while maintaining high storage efficiency. Specifically, we introduce a plug-and-play token redundancy reduction module delicately designed for PET. This module refines tokens from the self-attention layer using an adapter to learn the accurate similarity between tokens and cuts off the tokens through a fully-differentiable token merging strategy, which uses a straight-through estimator for optimal token reduction. Experimental results prove that our FPET achieves faster inference and higher memory efficiency than the pre-trained backbone while keeping competitive performance on par with state-of-the-art PET methods. The code is available at https://github.com/kyk120/fpet. English IEEE COMPUTER SOC Faster Parameter-Efficient Tuning with Token Redundancy Reduction Conference 10.1109/CVPR52734.2025.02810 1 2025 Conference on Computer Vision and Pattern Recognition-CVPR-Annual, pp.30189 - 30198 2025 Conference on Computer Vision and Pattern Recognition-CVPR-Annual 30189 30198 US Nashville, TN 2025-06-10 2025 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR) 001601181100570 2-s2.0-105017072261