Answers are grounded in this paper and cite their sources.
SERAPIS-E is a modular plateau model in which the plateau, the vacuum at ρ = e^(2πi/3) and a one-loop string tilt all come from the Eisenstein series E(τ,2) [1]. We report a full audit of its derivation chain, with every step recomputed under a pre-registered lock. The core arithmetic stands. Two input errors in the reheating chain are corrected: the inflaton mass at ρ is 3.17×10^13 GeV, not 2.2×10^13, and the inflaton couplings were normalised with the wrong Kähler weight. With the MSSM g_*=228.75 the reheating temperature becomes T_R=5.5×10^8–1.65×10^10 GeV and N_*=50.07–51.22. The audited range is n_s=0.97166–0.97596 and r=(4.39–4.90)×10^-4, partly outside the band we froze before the audit. A leptogenesis-fixed reheating temperature then gives a point prediction, n_s=0.97439, r=4.82×10^-4, α_s=-9.98×10^-4, which lies inside the frozen band. A fast slow-roll code agrees with exact Mukhanov–Sasaki spectra to |Δ n_s|≤1.1×10^-6 over the whole range. We measure how each input moves the observables and turn the result into a forecast. If T_R is known, Simons Observatory plus SPHEREx would fix the string combination Im S Im ρ to ±14%. Two side results are reported: one postulate passes (a gravitino-mass window) and one is killed (a running test separating the model's two vacua). The audited model inherits the 2.4σ preference of DESI DR2 for evolving dark energy over ΛCDM. All results after the audit are post hoc and are scored only on future data. Status of the claims. Lattice sums, BHK tilt, tadpole and vacuum curvature: computed (confirmed). Inflaton mass at ρ (3.17e13 GeV) and Kähler weight of the couplings (n = 1): computed corrections. Reheating window and N*: computed with the MSSM g*. Audited range and point prediction: post hoc, labelled postulate; the frozen band (15 Nov 2026) is unchanged and is still what gets scored. String-meter forecast and gravitino window: postulates that passed pre-stated checks. Vacuum discriminant via κ: killed. Late-universe Δχ² = −7.98: computed.
No takes yet. Share an insight, caveat, or question.
Danny Eldridge (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: