English(EN) Learning Dynamics of Logits Debiasing for Long-Tailed Semi-Supervised Learning

新DyTrim框架解决长尾半监督学习中的偏见问题

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-01 04:00

研究人员开发了DyTrim，一个新颖的动态剪枝框架，旨在解决长尾分布在半监督学习中的挑战。该方法从理论上刻画了logits去偏过程，揭示了伪标签如何偏向多数类。DyTrim通过对标记数据进行类别感知剪枝和对未标记数据进行基于置信度的软剪枝来重新分配梯度预算，旨在减少类别偏见并增强泛化能力。 AI

影响这项研究提供了一个理论框架和一种实用的方法，以改善在具有不平衡类别分布的真实世界数据集中的模型泛化能力。

排序理由该集群包含一篇详细介绍半监督学习新方法的论文。[lever_c_demoted from research: ic=1 ai=1.0]

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 30 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇详细介绍半监督学习新方法的论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, model release; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

arXiv cs.AI TIER_1 English(EN) · Yue Cheng, Jiajun Zhang, Xiaohui Gao, Weiwei Xing, Zhanxing Zhu · 2026-09-01 04:00

长尾半监督学习中 Logits 偏见消除的学习动力学

arXiv:2608.30699v1 Announce Type: cross Abstract: Long-tailed distributions are prevalent in real-world semi-supervised learning (SSL), where pseudo-labels tend to favor majority classes, leading to degraded generalization. While many long-tailed semi-supervised learning (LTSSL) …