Speaker
Description
Searching for the Quantum Chromodynamics (QCD) critical point remains a major objective of relativistic heavy-ion collision experiments. Conventional fluctuation analyses primarily rely on ratios of cumulants of conserved charges, which can exhibit sign changes, cancellations among different orders, and sensitivity to competing non-critical backgrounds. In this work, we develop an information-theoretic framework based on the Kullback--Leibler divergence, $D_{\mathrm{KL}}(P \parallel Q)$, as an integrated and sign-definite measure of the deviation of an experimental multiplicity distribution from a reference baseline. Following the Maximum Entropy principle, discrete net-proton multiplicity distributions $P(N)$ are reconstructed from the efficiency-corrected cumulants $C_1$--$C_4$ measured by the STAR Collaboration in Au+Au collisions over $\sqrt{s_{NN}} = 7.7$--$200$ GeV. To investigate the sensitivity of the information-theoretic signal to non-critical physics, the experimental distributions are compared with an uncorrelated Skellam baseline and with several Hadron Resonance Gas (HRG) formulations, including Ideal HRG and Excluded-Volume HRG (EV-HRG), with and without exact baryon number conservation (BNC). The resulting beam-energy dependence of $D_{\mathrm{KL}}$ exhibits a pronounced non-monotonic structure, with a maximum in the intermediate beam-energy region. Importantly, this non-monotonic behavior persists across the different HRG baselines considered, indicating that the observed deviation cannot be accounted for by repulsive interactions, resonance decays, or global baryon number conservation alone. These results demonstrate the potential of relative entropy as a complementary information-theoretic observable for characterizing non-Gaussian fluctuations and identifying beam-energy regions exhibiting enhanced deviations from non-critical baseline expectations, thereby providing a novel approach in the search for the QCD critical point.