用似然比设计更可靠的模型拒答机制，尤其适合测试数据分布变化时。
Know When to Abstain: Optimal Selective Classification with Likelihood Ratios
- 基于似然比设计拒答策略，理论依据更强。
- 在分布偏移场景下，性能优于现有方法。
- 适用于视觉、语言任务，尤其适合不确定预测的场景。
选择性分类通过允许模型在不确定时拒绝预测来提升预测可靠性。本文从奈曼-皮尔逊引理出发，将最优拒答规则重新诠释为似然比检验，不仅统一了多种后处理拒答方法的行为，还提出了新的方法。研究重点是协变量偏移场景——测试时输入分布与训练时不同，这一现实且具有挑战性的场景在选择性分类中仍较少被探讨。我们在多种视觉与语言任务（包括监督学习和视觉-语言模型）上评估所提方法，结果表明，受奈曼-皮尔逊启发的方法持续优于现有基线，证明似然比驱动的拒答在协变量偏移下具备稳健性。代码已公开于 https://github.com/clear-nus/sc-likelihood-ratios。
原文摘要 · Abstract (English)
Selective classification enhances the reliability of predictive models by allowing them to abstain from making uncertain predictions. In this work, we revisit the design of optimal selection functions through the lens of the Neyman--Pearson lemma, a classical result in statistics that characterizes the optimal rejection rule as a likelihood ratio test. We show that this perspective not only unifies the behavior of several post-hoc selection baselines, but also motivates new approaches to selective classification which we propose here. A central focus of our work is the setting of covariate shift, where the input distribution at test time differs from that at training. This realistic and challenging scenario remains relatively underexplored in the context of selective classification. We evaluate our proposed methods across a range of vision and language tasks, including both supervised learning and vision-language models. Our experiments demonstrate that our Neyman--Pearson-informed methods consistently outperform existing baselines, indicating that likelihood ratio-based selection offers a robust mechanism for improving selective classification under covariate shifts. Our code is publicly available at https://github.com/clear-nus/sc-likelihood-ratios.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。