Tackling the challenges of galaxy-dark matter connection modeling and new insights into secondary (assembly) biases

Modeling galaxy-dark matter connection is essential in deriving unbiased cosmological constraints from galaxy clustering observations. We show that a more physically motivated galaxy-dark matter incorporating secondary (assembly) biases results in more accurate predictions of galaxy clustering, yields novel insights into effects such as baryonic feedback, and significantly reduces the tension in galaxy lensing. We further present progress and opportunities in building a multi-tracer galaxy-dark matter connection framework that is rich in features and highly efficient, enabling more robust cosmological analyses with upcoming DESI data.