Reproduce “Sequential auditing of connected forecast errors under partial observations” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Yuanbo Zhou, Junjie Xiong

The authors present an audit method that retrieves outcomes only as needed to decide whether three connected locations fall on the same side of their forecast intervals. At 30% masking, MOMS uses 19.6%–38.5% fewer queries than screened random, while shortest witness finds more positive cases early in tested METR-LA localized masks; these interval violations are not verified faults.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.