MERIT：缓解生成式XMC中的暴露偏差以建模用户兴趣倾向

AI 摘要

MERIT 框架旨在缓解生成式极端多标签分类（XMC）中的暴露偏差，用于用户兴趣倾向建模。通过置换不变的多目标损失和混合金标签与挖掘的难负样本，该方法在包含 25 万以上兴趣类别的电商数据集上，将全局召回率提升至少 11.9%，平均 Hit@k 提升 6.1%，并在生产 A/B 测试中实现用户转化率提升 0.26%。该研究由 arXiv 发布，属于学术研究。

正文节选

MERIT: Mitigating Exposure Bias in Generative XMC for User-Interest Propensity Modeling Abstract Matching users to interest categories at scale is central to personalized shopping, but the task is challenging in large e-commerce platforms, where label spaces continually evolve and user-interest signals are sparse and long-tailed. Autoregressive language models are appealing because their world knowledge and semantic priors over descriptors generalize across extreme label spaces and accommodate m

发布时间：2026-09-01 12:00

抓取时间：2026-09-01 12:16

来源机构：arXiv