A model-independent test of pre-recombination New Physics: Machine Learning based estimate of the Sound Horizon from Gravitational Wave Standard Sirens and the Baryon Acoustic Oscillation Angular Scale

In any cosmological model where spacetime is described by a pseudo-Riemannian manifold, photons propagate along null geodesics, and their number is conserved, upcoming Gravitational Wave (GW) observations can be combined with measurements of the Baryon Acoustic Oscillation (BAO) angular scale to provide model-independent estimates of the sound horizon at the baryon-drag epoch. By focusing on the accuracy expected from forthcoming surveys such as LISA GW standard sirens and DESI or Euclid angular BAO measurements, we forecast a relative precision of ₑ_ ₃ /r ₃ 1. 5\% within the redshift range z 1. This approach will offer a unique model-independent measure of a fundamental scale characterizing the early universe, which is competitive with model-dependent values inferred within specific theoretical frameworks. These measurements can serve as a consistency test for CDM, potentially clarifying the nature of the Hubble tension and confirming or ruling out new physics prior to recombination with a statistical significance of 4.