arXiv:2608.11287v1 Announce Type: new Abstract: Long-tailed classification poses a reliability challenge because models trained on imbalanced data are unevenly reliable across frequent and underrepresented classes. Whil
Sign in with a social account
在真实世界的图像分类任务中，类别分布通常呈长尾形态：少数类别拥有大量样本，而绝大多数类别仅有少量样本。直接在这种不平衡数据上训练模型，会导致模型对样本充足的"头部类"过度自信，对样本稀缺的"尾部类"则表现不稳定。CLEAR 的核心思路是将分类任务拆解到类级别——为不同类别训练或聚合专门的"专家"模块，并配合一种结构化采样策略来平衡训练过程中的类别暴露频率。不过，摘要中关于专家聚合的具体机制（如是否采用路由网络、加权投票或层次化集成）以及结构化采样的实现细节，原文摘要未充分展开，具体技术方案需查阅全文。
从摘要透露的信息看，CLEAR 的出发点是将"可靠性"作为长尾分类的核心评估维度，而非仅仅追求整体准确率。这一视角的转变本身值得关注：传统长尾分类方法常以平均精度或各类别精度的简单平均为指标，而可靠性视角更强调模型在不同类别间的表现一致性。作为 arXiv 预印本，该研究尚未经过正式同行评议，其声称的效果需要等待后续实验数据与复现结果来验证。论文的实验规模、数据集选择、对比基线等关键信息在摘要中均未提供，因此目前只能将其视为一项方法性提案。
如果 CLEAR 的类级专家聚合策略确实能在不显著增加推理成本的前提下改善尾部类可靠性，它对需要处理长尾分布的实际场景——如罕见物种识别、医疗影像中的少见病筛查、工业缺陷检测等——可能具有参考价值。但需要保持审慎：类级专家方法通常面临参数量随类别数线性增长的扩展性问题，且尾部类样本过少时，即便有结构化采样辅助，专家模块本身也可能欠拟合。此外，该研究与同日发布的其他 arXiv 论文（如关于多智能体治理、手语翻译、提示优化等）并无直接关联，属于独立的计算机视觉方向探索。
arXiv:2608.11287v1 Announce Type: new Abstract: Long-tailed classification poses a reliability challenge because models trained on imbalanced data are unevenly reliable across frequent and underrepresented classes. While existing methods address imbalance through re-balancing, adjustment, representation learning, or multi-expert modeling, they rarely estimate which expert should be trusted for each class. This paper proposes CLEAR (Class-wise reLiability-aware Expert Aggregation for long-tailed Recognition), a modular ensemble framework for long-tailed classification. CLEAR generates diverse experts through threshold-based structured sampling while preserving the full label space, then estimates a class-wise trust score for each expert using a smoothed class-wise precision formulation. During inference, expert predictions are combined through class-wise generalized product-of-experts aggregation, allowing different experts to be emphasized for different classes. Experiments on CIFAR-100-LT, ImageNet-LT, and Places-LT across multiple backbones show that CLEAR achieves competitive overall accuracy and particularly strong few-shot performance. These results support class-wise expert reliability as a useful design principle for long-tailed ensemble learning.