Raul Jimenez, Carlos Peña Garay, Fergus Simpson, Licia Verde
Cosmological data have reached the precision needed to turn the neutrino mass ordering from a weak Bayesian preference into a decisive model-selection test. We compute the evidence for the Normal and Inverted Hierarchies by combining DESI DR2 clustering with NuFIT oscillation data. In baseline CDM, DESI DR2 plus Planck CamSpec gives at 95\% confidence, close to the normal-ordering floor, , but well below the inverted-ordering minimum, . Thus the inverted hierarchy lies in the tail of the cosmological likelihood. The Bayes factor exceeds even for a conservative reference prior, and remains strong, , in baseline-model extensions. We show that this result is robust to the choice between a reference prior and a physically motivated logarithmic hierarchical prior, marking the transition from {\em prior-sensitive evidence} to {\em likelihood-dominated exclusion} of the inverted hierarchy within standard cosmology. Embedding these priors in the two-dimensional design space of measure (logarithmic versus linear in mass) and structure (hierarchical versus non-hierarchical), we find that all four prior constructions give decisive evidence under DESI DR2, with residual prior dependence governed mainly by the measure -- a factor in -- rather than by the hierarchy assumption. At the prior-family level, the evidence favors the SJPV prior predictive over HS by a Bayes factor above across each matched-support variation tested. The favored normal ordering pushes the effective Majorana mass to the few-meV regime, with median and 95\% credible interval , below the inverted-ordering target for upcoming neutrinoless double-beta decay experiments.
This paper claims "decisive" Bayesian evidence for the Normal Hierarchy (NH) of neutrino masses over the Inverted Hierarchy (IH), driven by the DESI DR2 BAO + Planck CamSpec cosmological constraint Σmν < 0.0642 eV (95% CL). Since the IH floor is ~0.099 eV, the entire IH parameter space now lies in the tail of the cosmological likelihood. The central result is a Bayes factor K > 460 under the conservative HS reference prior and K > 10³ under the SJPV hierarchical prior, both crossing Jeffrey's "decisive" threshold.
The paper's most intellectually distinctive contribution is the systematic decomposition of prior dependence into a 2×2 design space (logarithmic vs. linear measure × hierarchical vs. non-hierarchical structure). This disentangles two axes that were previously conflated in the SJPV vs. HS debate, revealing that the measure choice (factor ~10 in K) dominates over the hierarchical structure assumption (factor ~2). This is a genuinely clarifying methodological insight that resolves a longstanding source of confusion in the literature.
Strengths: The analytical framework is well-constructed. The truncated Gaussian approximation to the marginalized posterior is validated against KDE reconstruction and Feldman-Cousins profiling. The historical reconstruction of K across two decades of data provides valuable context and demonstrates the transition from prior-dominated to likelihood-dominated inference. The 2×2 prior design space is a clean conceptual framework, and the evidence integrals are validated against full three-dimensional calculations (|Δln Z| < 0.006).
Concerns:
1. The cosmological likelihood is taken entirely at face value. The authors explicitly state they are "not concerned on the robustness of the DESI neutrino mass results." This is a significant caveat, as the DESI DR2 constraint pushes Σmν close to or below the physical floor—a regime where systematic effects, likelihood approximations, and prior volume effects in the original MCMC chains deserve scrutiny. The paper essentially conditions all conclusions on the validity of someone else's analysis.
2. The truncated Gaussian with μ₀ = -0.036 eV is a pragmatic but potentially fragile parametrization. While the authors check alternative functional forms (exponential, half-Gaussian), the claim that these differ by <O(10%) in K is important but deserves more detailed presentation, given that the entire inference hinges on the shape of this likelihood in the 0.06-0.10 eV region.
3. The w₀waCDM robustness check yields K > 40 (HS), which is "strong" but no longer "decisive" by Jeffrey's scale. The paper acknowledges this but frames it somewhat asymmetrically—the title claims "decisive" evidence, yet the most physically motivated cosmological extension reduces the evidence below that threshold for the conservative prior.
4. The one-dimensional reduction of the evidence integral (analytically marginalizing the two oscillation-constrained directions) is validated but the full three-dimensional treatment is presented as verification rather than the primary calculation, which raises questions about whether subtleties in the oscillation covariance structure could matter.
The implications are significant across multiple fields:
The paper is extremely timely, appearing shortly after DESI DR2 results that produced the tightest cosmological neutrino mass bounds to date. It directly addresses the question on many physicists' minds: what do these new bounds mean for the mass hierarchy? The paper also arrives at a moment when JUNO has released first results, and the interplay between cosmological and laboratory determinations of the ordering is a live question.
Key Strengths:
Notable Limitations:
This is a timely, well-executed Bayesian analysis that synthesizes DESI DR2 constraints with oscillation data to make a strong claim about the neutrino mass hierarchy. The 2×2 prior design space is a genuine methodological contribution. However, the conditional nature of the inference (entirely dependent on the validity of the cosmological constraints) and the somewhat overstated framing relative to the w₀waCDM results temper the impact. The paper will be widely cited and discussed, but the claim of "decisive" evidence will remain debated until the cosmological constraints themselves are independently validated and confirmed to be robust to systematic effects.
Generated Jun 18, 2026
Paper 1 addresses a fundamental open question in the Standard Model of particle physics: the neutrino mass hierarchy. By providing decisive cosmological evidence for the Normal Hierarchy, this result has profound implications for both cosmology and particle physics, directly impacting the viability and design of major upcoming neutrinoless double-beta decay experiments. While Paper 2 offers highly precise astrophysical measurements crucial for baryonic feedback modeling, Paper 1's resolution of a core fundamental physics problem gives it significantly broader scientific impact and transformative potential across multiple disciplines.
Paper 2 introduces Lumina, a groundbreaking large-volume radiation-hydrodynamic simulation that self-consistently models both hydrogen and helium reionization with unprecedented scale and resolution. This represents a major computational and methodological advance with broad applicability across cosmology, galaxy formation, IGM physics, and observational survey interpretation. It will serve as a community resource enabling numerous follow-up studies. Paper 1, while timely and significant in establishing cosmological evidence for the normal neutrino mass hierarchy using DESI DR2 data, is more narrowly focused on a specific measurement whose conclusion (normal ordering preference) has been building incrementally and may be superseded by direct laboratory measurements.
Paper 1 addresses a fundamental open question in particle physics and cosmology: the neutrino mass hierarchy. By providing decisive statistical evidence for the Normal Hierarchy using recent DESI DR2 data, it heavily impacts the Standard Model and dictates the viability of future neutrinoless double-beta decay experiments. While Paper 2 offers an innovative 3.2-sigma observational constraint on void baryons using FRBs, resolving the neutrino mass hierarchy represents a more transformative breakthrough with broader, field-defining implications for theoretical and experimental physics.
Paper 1 likely has higher impact: it claims decisive, likelihood-dominated cosmological exclusion of the inverted neutrino mass hierarchy using cutting-edge DESI DR2 + Planck + oscillation data, addressing a long-standing fundamental question with broad implications for particle physics, cosmology, and neutrinoless double-beta decay prospects. The result is timely, high-profile, and immediately constrains multiple research programs. Paper 2 is methodologically innovative (two-loop EFT cosmic shear) with strong future utility, but its primary outcomes (competitive S8 constraints, reduced tension) are more incremental and depend on adoption by the community.
Paper 2 proposes a concrete, actionable observational program combining Rubin and DESI that could definitively confirm or refute dynamical dark energy at >5σ within one year—potentially transforming our understanding of dark energy. Its practical survey design, systematics mitigation strategy, and broad applicability to future experiments give it higher impact potential. Paper 1, while presenting a strong cosmological result on neutrino mass ordering, largely confirms an already-favored hierarchy using existing data (DESI DR2 + Planck), making it more incremental. Paper 2's potential to resolve the ΛCDM tension and guide next-generation experiments gives it broader transformative impact.
Paper 1 likely has higher overall scientific impact because it delivers a massive, reusable public infrastructure (2.3+ PB) enabling many independent studies across cosmology, galaxy formation, large-scale structure, lensing/lightcones, neutrinos, and dark-matter models. Its breadth and longevity as a community dataset/resource can generate sustained, wide citation and downstream discoveries. Paper 2 is highly timely and potentially headline-making (decisive NH vs IH from DESI+Planck), but its impact is narrower, more model/assumption-contingent, and may be superseded as datasets/analyses evolve. Overall: broader, enabling impact favors Paper 1.
Paper 1 delivers a landmark result using actual observational data (DESI DR2) to decisively resolve a major open question in fundamental physics: the neutrino mass hierarchy. By establishing the Normal Hierarchy and providing concrete mass targets for future neutrinoless double-beta decay experiments, it has immediate, profound impacts on both cosmology and experimental particle physics. In contrast, Paper 2, while methodologically innovative in applying topological data analysis, is currently a proof-of-concept tested on simulations. Paper 1's definitive real-world measurement ensures a significantly higher and broader scientific impact.
Paper 1 likely has higher overall scientific impact because a 2.3+ PB public hydrodynamical simulation suite is a durable, widely reusable community resource: it can enable many downstream studies (galaxy formation, large-scale structure, lensing, clusters, neutrinos, dark-matter models) and supports cross-field applications and survey analysis pipelines. Its impact breadth and longevity are high, and the data-access infrastructure increases adoption. Paper 2 is timely and potentially transformative for neutrino physics, but its influence is narrower and more contingent on cosmological-model assumptions and future cross-checks.
Paper 2 presents multiple first-ever detections (velocity-galaxy cross-correlation for ELGs and QSOs, velocity-velocity auto-correlation for LRGs) and the highest cumulative kSZ detection to date (20.8σ). It opens a new observational window for constraining primordial non-Gaussianity from the velocity field and demonstrates that kSZ-based measurements can surpass galaxy clustering on the largest scales, establishing methodology critical for future CMB experiments like Simons Observatory. Paper 1, while presenting a strong cosmological result on neutrino mass ordering, is largely a Bayesian reanalysis of existing DESI DR2 data with known physics, and its conclusions depend on the ΛCDM framework. Paper 2's novel measurements and broader methodological impact across cosmology give it higher potential.
Paper 2 addresses a fundamental, long-standing question in particle physics and cosmology: the neutrino mass hierarchy. By providing decisive evidence for the Normal Hierarchy using recent DESI DR2 data, it profoundly impacts theoretical physics and the design of upcoming neutrinoless double-beta decay experiments. While Paper 1 offers valuable statistical methodology improvements for covariance estimation, Paper 2's breakthrough results on foundational physics parameters give it a vastly broader scope, higher timeliness, and significantly greater potential scientific impact across multiple physics disciplines.