Triple Expert Learning from Noisy Labels for Semi-Supervised Vision Foundation Model Adaptation
面向半监督视觉基础模型适配的带噪标签三重专家学习
机构 * The University of Warwick(华威大学) ; Wuhan University(武汉大学)
专题命中 指令微调 :foundation model(title,abstract);分类 cs.AI
AI总结 TriNoL框架通过将未标记样本按伪标签置信度分配给三个LoRA专家,实现半监督视觉基础模型适配,提升了对带噪监督的鲁棒性且训练成本较低。
Comments Accepted for publication at the British Machine Vision Conference (BMVC) 2026. Official list of accepted papers: https://bmvc2026.bmva.org/programme/accepted_papers/