Abstract:Exploiting unlabeled data through semi-supervised learning (SSL) or leveraging pre-trained models via fine-tuning are two prevailing paradigms for addressing label-scarce scenarios. Recently, growing attention has been given to combining fine-tuning of pre-trained vision-language models (VLMs) with SSL, forming the emerging paradigm of semi-supervised fine-tuning. However, existing methods often suffer from model bias and hyperparameter sensitivity, due to reliance on prediction consistency or pre-defined confidence thresholds. To address these limitations, we propose a simple yet effective plug-and-play methodology named $\underline{\textbf{Bi-Co}}$nsistency-$\underline{\textbf{G}}$uided Self-Training (Bi-CoG), which assigns high-quality and low-bias pseudo-labels, by simultaneously exploiting inter-model and intra-model consistency, along with an error-aware dynamic pseudo-label assignment strategy. Both theoretical analysis and extensive experiments over 14 datasets demonstrate the effectiveness of Bi-CoG, which consistently and significantly improves the performance of existing methods.
| Comments: | Accepted by IJCAI 2026 |
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2510.20477 [cs.LG] |
| (or arXiv:2510.20477v3 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2510.20477 arXiv-issued DOI via DataCite |
From: Rui Zhu [view email]
[v1]
Thu, 23 Oct 2025 12:16:41 UTC (503 KB)
[v2]
Sun, 10 May 2026 16:16:46 UTC (481 KB)
[v3]
Sat, 23 May 2026 08:43:25 UTC (425 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。