边际覆盖率无法保证零样本VLM在偏移下的类别条件安全性
原标题：Does Marginal Coverage Guarantee Class-Conditional Safety for Zero-Shot VLMs Under Shift?
AI 摘要
该研究审计了零样本视觉语言模型（如CLIP、OpenCLIP、SigLIP）在分布偏移下使用分裂共形预测的类别条件安全性。研究发现，边际覆盖率看似较高（如ImageNet-Sketch上约0.86），但最差类别的覆盖率可降至0，且10-12%的类别低于有限样本零下限。源域诊断无法预测失败，源端Mondrian校准不迁移，目标端类别校准虽提升尾部但需每类标签且集合大小代价高。结论是边际共形覆盖率应视为平均可靠性统计，而非类别尾部的安全保证。
正文节选
Does Marginal Coverage Guarantee Class-Conditional Safety for Zero-Shot VLMs Under Shift? Abstract Split-conformal prediction provides marginal coverage under exchangeability and is increasingly used as an abstention layer for zero-shot vision-language models (VLMs). We audit this practice under deployment shift for CLIP, OpenCLIP, and SigLIP across ImageNet and non-ImageNet settings. Marginal coverage can remain relatively high while class-conditional tail coverage collapses: on ImageNet-Sketch