LoFT: Parameter-Efficient Fine-Tuning for Long-tailed Semi-Supervised Learning in Open-World Scenarios

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Huang, Zhiyuan, Chen, Jiahao, Su, Bing
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913012492795904
author Huang, Zhiyuan
Chen, Jiahao
Su, Bing
author_facet Huang, Zhiyuan
Chen, Jiahao
Su, Bing
contents Long-tailed semi-supervised learning (LTSSL) presents a formidable challenge where models must overcome the scarcity of tail samples while mitigating the noise from unreliable pseudo-labels. Most prior LTSSL methods are designed to train models from scratch, which often leads to issues such as overconfidence and low-quality pseudo-labels. To address this problem, we first theoretically prove that utilizing a foundation model significantly reduces the hypothesis complexity, which tightens the generalization bound and in turn minimizes the Balanced Posterior Error (BPE). Furthermore, we demonstrate that the feature compactness of foundation models strictly compresses the acceptance region for outliers, providing a geometric guarantee for robustness. Motivated by these theoretical insights, we extend LTSSL into the foundation model fine-tuning paradigm and propose a novel framework: LoFT (Long-tailed semi-supervised learning via parameter-efficient Fine-Tuning). Furthermore, we explore a more practical setting by investigating semi-supervised learning under open-world conditions, where the unlabeled data may include out-of-distribution (OOD) samples.To handle this problem, we propose LoFT-OW (LoFT under Open-World scenarios) to improve the discriminative ability. Experimental results on multiple benchmarks demonstrate that our method achieves superior performance. Code is available: https://github.com/games-liker/LoFT
format Preprint
id arxiv_https___arxiv_org_abs_2509_09926
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle LoFT: Parameter-Efficient Fine-Tuning for Long-tailed Semi-Supervised Learning in Open-World Scenarios
Huang, Zhiyuan
Chen, Jiahao
Su, Bing
Machine Learning
Computer Vision and Pattern Recognition
Long-tailed semi-supervised learning (LTSSL) presents a formidable challenge where models must overcome the scarcity of tail samples while mitigating the noise from unreliable pseudo-labels. Most prior LTSSL methods are designed to train models from scratch, which often leads to issues such as overconfidence and low-quality pseudo-labels. To address this problem, we first theoretically prove that utilizing a foundation model significantly reduces the hypothesis complexity, which tightens the generalization bound and in turn minimizes the Balanced Posterior Error (BPE). Furthermore, we demonstrate that the feature compactness of foundation models strictly compresses the acceptance region for outliers, providing a geometric guarantee for robustness. Motivated by these theoretical insights, we extend LTSSL into the foundation model fine-tuning paradigm and propose a novel framework: LoFT (Long-tailed semi-supervised learning via parameter-efficient Fine-Tuning). Furthermore, we explore a more practical setting by investigating semi-supervised learning under open-world conditions, where the unlabeled data may include out-of-distribution (OOD) samples.To handle this problem, we propose LoFT-OW (LoFT under Open-World scenarios) to improve the discriminative ability. Experimental results on multiple benchmarks demonstrate that our method achieves superior performance. Code is available: https://github.com/games-liker/LoFT
title LoFT: Parameter-Efficient Fine-Tuning for Long-tailed Semi-Supervised Learning in Open-World Scenarios
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2509.09926