用三个专家分离伪标签可信度，提升半监督视觉模型适应鲁棒性。
Triple Expert Learning from Noisy Labels for Semi-Supervised Vision Foundation Model Adaptation
- 将伪标签按可信度分三类，分别对应三个LoRA专家
- 在ImageNet-1K上达到85.6%准确率，比基线高2.3个百分点
- 适合噪声数据多的半监督视觉模型微调场景
半监督视觉基础模型（VFM）适配通常冻结预训练主干网络，仅更新轻量级模块如LoRA。但伪标签可靠性参差不齐，单一LoRA适配器需在同一低秩空间中吸收可靠、模糊和噪声梯度，易受伪标签噪声影响。本文提出TriNoL——一种从噪声标签中进行三专家学习的框架。TriNoL将未标注样本划分为三个置信度区域，并分配给三个LoRA专家：高置信度伪标签对应正向专家，中等置信度模糊样本对应对齐专家，低置信度噪声样本对应负向专家。模型主干保持冻结，仅更新LoRA专家与分类头。通过将不同可信度伪标签分流至专用适配路径，TriNoL在保持低训练成本的同时显著提升对噪声监督的鲁棒性。
原文摘要 · Abstract (English)
Semi-supervised adaptation of vision foundation models (VFMs) commonly freezes the pretrained backbone and updates lightweight modules such as LoRA. However, pseudo-labels have mixed reliability, and a single LoRA adapter must absorb reliable, ambiguous, and noisy gradients in the same low-rank space. This can make VFM adaptation sensitive to pseudo-label noise. We propose \textbf{TriNoL}, a \textbf{Tri}ple-expert learning framework from \textbf{No}isy \textbf{L}abels for semi-supervised VFM adaptation. TriNoL routes unlabeled samples into three confidence regions and assigns them to three LoRA experts: a Positive Expert for high-confidence pseudo-labels, an Alignment Expert for medium-confidence ambiguous samples, and a Negative Expert for low-confidence noisy samples. The VFM backbone remains frozen, and only the LoRA experts and classifier head are updated. By separating different pseudo-label reliability regions into specialized adaptation paths, TriNoL improves robustness to noisy supervision while keeping the training cost low.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。