Researchers connected with URI’s gravity group have contributed to a set of recent short-author papers that push forward several key parts of modern gravitational-wave science: modeling the final products of black hole mergers, accelerating waveform calculations with machine learning, and improving the way gravitational-wave detectors identify and interpret signals.
One recent paper, “Unified remnant models for aligned-spin, precessing, and eccentric binary black hole mergers”, by Tousif Islam, Digvijay Wadekar, and Gaurav Khanna, presents new analytic models for predicting the final mass, spin, luminosity, and recoil of merged black holes. The work combines thousands of numerical-relativity simulations with black-hole-perturbation-theory calculations, extending accuracy across regimes from comparable-mass binaries to extreme mass ratios. These models are intended for applications in gravitational-wave astronomy, astrophysical population studies, and cosmology.
A second paper, “Fast, accurate, and differentiable: a neural-network surrogate for NRSur7dq4 precessing binary black hole waveforms”, by Michael Pürrer, Ashwin Girish, Lucy M. Thomas, Scott E. Field, and Vijay Varma, develops a neural-network surrogate for precessing binary black hole waveforms. The model reproduces the NRSur7dq4 waveform family with high accuracy while running much faster on GPUs. Because the full waveform-to-likelihood pipeline is implemented in JAX and is differentiable, it opens the door to faster parameter estimation and gradient-based inference methods for gravitational-wave data analysis.
The group’s detector-analysis work is represented by “SGNAX: a unified matched-filter and excess-power pipeline for gravitational-wave detector characterization”, by Zach Yarbrough, Olivia Godwin, Derek Davis, and Gabriela González. SGNAX is an open-source pipeline that unifies two important approaches to detector characterization in a single streaming Python framework. It can process a full day of gravitational-wave strain data in minutes, supports both CPU and GPU execution, and is designed for offline and low-latency detector studies.
Together, these papers highlight the range of computational work behind gravitational-wave discovery. Accurate black hole remnant models help scientists interpret the astrophysical meaning of observed mergers. Fast waveform surrogates make large-scale inference more practical. Detector-characterization pipelines help separate real astrophysical signals from instrumental artifacts. Across theory, computation, and data analysis, URI-affiliated researchers continue to contribute tools that support the next generation of gravitational-wave astronomy.