发表机构
Columbia University; EleutherAI; UniverseTBD; Indian Institute of Science Education and Research, Thiruvananthapuram; AstroAI; Center for Astrophysics | Harvard & Smithsonian; Washington University in St. Louis（哥伦比亚大学; EleutherAI; UniverseTBD; 印度科学教育与研究学院（特里凡得琅）; AstroAI; 哈佛-史密森天体物理中心; 圣路易斯华盛顿大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出共形框架审计跨巡天天文基础模型表示，发现AstroPT嵌入迁移误差更低，并利用自适应共形预测识别可靠迁移与静默失败。
AI 中文摘要
天文基础模型越来越多地被用作跨巡天的可重用表示，但尚不清楚同一目标在不同仪器间是否保持一致表示，或者模型能否在迁移失败前发出警告。我们提出了一种共形框架，利用交叉匹配的 Legacy Survey–HSC、HSC–JWST 和 SDSS–HSC 目标来审计跨巡天表示。天文专用的 AstroPT 嵌入在可比覆盖率下显示出比通用视觉骨干更低的平移误差和更紧致的共形区域，而不同模型家族在不同目标上失败。然后，我们使用自适应共形预测来测试源嵌入是否包含关于其自身未来误差的信息。使用拟合到源嵌入的岭难度模型，我们发现该警告信号在测试的骨干中对于 Legacy→HSC 是一致的，对于 HSC→JWST 依赖于骨干，而对于 SDSS→HSC 则基本缺失。当信息有用时，自适应改进对最难目标的覆盖率。匹配对照分析进一步揭示了无法用星等、视大小或最近邻距离解释的静默失败。因此，我们表明共形不确定性提供了一种现实的方法来识别天文表示何时可靠迁移以及何时在无警告的情况下失败。
英文摘要
Astronomical foundation models are increasingly used as reusable representations across surveys, but it is unclear when the same object remains consistently represented across instruments or whether a model can warn us before transfer fails. We present a conformal framework for auditing cross-survey representations using cross-matched Legacy Survey--HSC, HSC--JWST, and SDSS--HSC objects. Astronomy-specific AstroPT embeddings show lower translation errors and tighter conformal regions than general-purpose visual backbones at comparable coverage, while different model families fail on different objects. We then use adaptive conformal prediction to test whether the source embedding contains information about its own future error. Using a ridge difficulty model fitted to source embeddings, we find that this warning signal is consistent across the tested backbones for Legacy$\rightarrow$HSC, backbone-dependent for HSC$\rightarrow$JWST, and largely absent for SDSS$\rightarrow$HSC. When informative, adaptation improves coverage for the hardest objects. Matched-control analysis further reveals silent failures not explained by magnitude, apparent size, or nearest-neighbor distance. Hence, we show that conformal uncertainty provides a realistic way to identify when astronomical representations transfer reliably and when they fail without warning.
Comments23 pages, 17 figures, 4 tables, accepted to the Interpretability for Discovery workshop at NeurIPS 2026
