AI and the World's Hard Problems · Physics · arXiv · 2026-10-22
Towards anomaly detection searches for new physics signatures including Higgs bosons with weakly supervised machine learning
The Higgs boson's universal coupling to mass makes it a natural anchor for anomaly detection searches for physics beyond the Standard Model at collider experiments. The HAXAD strategy combines machine-learning-based feature embedding, background estimation, and weakly supervised classification, and this work brings it toward the maturity needed for recorded collider data by introducing two new embedding strategies, an inference framework yielding signal-agnostic and signal-specific cross section limits, and a significantly expanded set of signal models. The improvements increase signal sensitivity over the original method, and benchmarked against an example cut-based search on the same final state, HAXAD matches or exceeds the best individual cut-based limits for a wide variety of signal models.