DESI Two-Point Clustering: Measurements, models, and cosmological interpretation

DESI is a wide-field spectroscopic survey mapping the three-dimensional distribution of galaxies and quasars to enable precision tests of cosmic expansion and structure growth. DESI's measurements in combination with external data offer tantalizing hints of deviations from the standard cosmological model, specifically related to the Einstein cosmological constant. Key part of these results are driven by baryon acoustic oscillations (BAO), but not all of the information is there; the full shape of the correlation function (or alternatively, power spectrum) carries complementary cosmological information. Extracting this information is more challenging: the complexity of both the measurements and the modeling increases substantially because we can no longer rely on linear physics alone, requiring careful control of observational systematics and robust treatment of non-linear bias, redshift-space distortions, and mode coupling. While full-shape analyses extract rich expansion and growth information across DESI's tracers and redshifts, they require complex modeling whose interpretation within the Bayesian framework is not straightforward—particularly due to projection effects. These projection effects can shift one-dimensional posteriors away from the best-fitting solution and bias inference. I will outline three complementary mitigation strategies: (i) simulation-calibrated priors, (ii) a frequentist profile-likelihood approach, and (iii) a Bayesian re-parameterisation that isolates degenerate directions. Applied to full-shape DESI data, each method reduces the impact of projection effects. I will summarize the current status of the full-shape analysis and close with an outlook on forthcoming checks and applications aimed at more precise and interpretable cosmological constraints.