谱共形风险控制：基于贝叶斯求积的分布无关尾部保证
Spectral Conformal Risk Control: Distribution-Free Tail Guarantees via Bayesian Quadrature
为视觉模型提供分布无关的尾部风险（CVaR 式）控制手段：只需黑盒访问已训练模型，即可在分割、多标签与零样本分类等任务上构造具有有限样本保证的预测集，适合医疗影像、自动驾驶等对罕见高代价错误敏感的场景。
原文依据 BQ-SRC views conformal prediction through a Bayesian-quadrature lens and replaces mean-risk control with a flexible family of risk-averse criteria
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论现代视觉系统部署在偶发灾难性失败比平均精度更重要的场景中——例如医学影像、自动驾驶和安全监控。尽管共形预测提供了分布无关的不确定性保证，但现有方法大多只控制平均误差，难以针对罕见但高代价的错误进行调节。我们提出贝叶斯求积谱风险控制（BQ-SRC），这是一个以分布无关方式控制尾部聚焦风险（如条件风险价值（CVaR）式目标）的通用框架。BQ-SRC 从贝叶斯求积的视角审视共形预测，用一族灵活的风险规避准则替代平均风险控制，同时保持对已训练模型的相同黑盒访问方式。一种二项检验方案降低了先前方法的蒙特卡洛保守性，从而在不牺牲保证的前提下得到更紧的集合。我们在多种视觉任务上评估 BQ-SRC，包括合成回归、闭集与零样本图像分类、多标签分类以及语义分割。在这些设置中，BQ-SRC 始终维持有限样本风险保证，并且往往比现有的共形与风险控制基线产生更小或其他方面信息更丰富的预测集，有时以适度的效率损失换取更强的尾部风险控制。代码将公开于 https://github.com/MohammadMahdiKazemi/BQ_SRC。
英文原文摘要 1484 字符
Modern vision systems are deployed in settings where occasional catastrophic failures matter more than average accuracy--for example in medical imaging, autonomous driving, and safety monitoring. While conformal prediction gives distribution-free uncertainty guarantees, most existing methods only control mean error and are hard to tune toward rare but high-cost mistakes. We propose Bayesian-Quadrature Spectral Risk Control (BQ-SRC), a general framework for controlling tail-focused risks (such as conditional value at risk (CVaR)-style objectives) in a distribution-free way. BQ-SRC views conformal prediction through a Bayesian-quadrature lens and replaces mean-risk control with a flexible family of risk-averse criteria, while keeping the same black-box access to a trained model. A binomial testing scheme reduces the Monte Carlo conservatism of prior approaches, leading to tighter sets without sacrificing guarantees. We evaluate BQ-SRC across diverse vision tasks, including synthetic regression, closed-set and zero-shot image classification, multilabel classification, and semantic segmentation. Across these settings, BQ-SRC consistently maintains finite-sample risk guarantees and often yields smaller or otherwise more informative prediction sets than existing conformal and risk-controlling baselines, sometimes trading a modest amount of efficiency for stronger tail-risk control. The code will be publicly available at https://github.com/MohammadMahdiKazemi/BQ_SRC.