{"slug": "clear-class-wise-expert-aggregation-with-structured-sampling-for-long-tailed", "title": "CLEAR: Class-wise Expert Aggregation with Structured Sampling for Long-Tailed Classification", "summary": "Researchers propose CLEAR (Class-wise reLiability-aware Expert Aggregation for long-tailed Recognition), a modular ensemble framework for long-tailed classification that estimates class-wise trust scores for experts and combines predictions via class-wise generalized product-of-experts aggregation. Experiments on CIFAR-100-LT, ImageNet-LT, and Places-LT show competitive overall accuracy and strong few-shot performance.", "body_md": "arXiv:2608.11287v1 Announce Type: new\nAbstract: 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.", "url": "https://wpnews.pro/news/clear-class-wise-expert-aggregation-with-structured-sampling-for-long-tailed", "canonical_source": "https://arxiv.org/abs/2608.11287", "published_at": "2026-08-13 04:00:00+00:00", "updated_at": "2026-08-13 04:12:42.502998+00:00", "lang": "en", "topics": ["machine-learning", "artificial-intelligence"], "entities": ["CLEAR", "CIFAR-100-LT", "ImageNet-LT", "Places-LT"], "alternates": {"html": "https://wpnews.pro/news/clear-class-wise-expert-aggregation-with-structured-sampling-for-long-tailed", "markdown": "https://wpnews.pro/news/clear-class-wise-expert-aggregation-with-structured-sampling-for-long-tailed.md", "text": "https://wpnews.pro/news/clear-class-wise-expert-aggregation-with-structured-sampling-for-long-tailed.txt", "jsonld": "https://wpnews.pro/news/clear-class-wise-expert-aggregation-with-structured-sampling-for-long-tailed.jsonld"}}