{"id":"5b844813-e01f-4bbe-ad4e-8b6728ffa16f","arxiv_id":"2607.06862","paper_version":1,"verdict":"CONDITIONAL","confidence":"UNKNOWN","novelty_score":4.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":5,"one_line_summary":"GLoBES simulations show that DUNE can probe quasi-Dirac sterile-neutrino mixing down to θ_sa ≈ 0.05 with uniform coverage, while current NOvA+T2K data constrains θ_sa ≳ 0.08.","lead":"This paper forecasts how well current (NOvA, T2K) and future (DUNE) long-baseline neutrino experiments can detect quasi-Dirac sterile neutrinos — neutrino pairs split by a tiny lepton-number-violating mass. A smart generalist might read it to understand what DUNE will be able to say about whether neutrinos are Dirac or Majorana particles.","discovery_kind":"extension","skeptic_critique":{"model":"glm-5.2","headline":"The flavor-universal θ_sa assumption collapses 6 independent mixing angles to 1, making the CP-suppression claim (Eq. 43) and all sensitivity contours special-case results rather than general properties of the quasi-Dirac framework.","rationale":"The reader's verdict of CONDITIONAL is appropriate. The flavor-universal θ_sa assumption is genuinely load-bearing: it underpins the analytic CP-suppression result (Eq. 43), the structure of the sensitivity contours, and the degeneracy analysis. The authors acknowledge this only in the Conclusion as a future direction, which is insufficient given that all quantitative results depend on it. However, the CONDITIONAL verdict already captures this uncertainty. The DUNE forecast remains the paper's strongest and most credible contribution — DUNE's long baseline and broad energy coverage wash out the degeneracy structures that plague NOvA/T2K, making the θ_sa ≈ 0.05 sensitivity claim more robust to the simplification than the NOvA/T2K bounds. The secondary concern about lack of marginalization over standard parameters (especially θ₂₃) would further weaken the NOvA/T2K constraints but is less impactful for DUNE. The paper's methodology (GLoBES simulation, systematic treatment, analytic derivations) is sound within its stated assumptions. The distinction between 'data from NOvA/T2K' (abstract) and simulated forecasts should be sharpened, but this is a framing issue rather than a scientific error. No verdict adjustment needed beyond what the reader already assigned.","tokens_in":23681,"tokens_out":3190,"duration_ms":175485,"concrete_test":"Re-run the DUNE χ² sensitivity scan with a non-universal mixing ansatz, e.g., θ₁₄ = θ₁₅ = 0.01 (KATRIN-constrained) while θ₂₄ = θ₂₅ = θ₃₄ = θ₃₅ = θ_sa, for Δm²₅₄ = 0.01 eV². If the 3σ exclusion boundary in θ_sa shifts by more than a factor of 2 relative to the universal case, the flavor-universal assumption is materially inflating the sensitivity claim. Additionally, re-derive Eq. 43 without assuming equal angles to check whether the c_sa factor generalizes or fragments into angle-dependent terms.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The reader correctly identifies the load-bearing concern. All quantitative results — sensitivity contours, degeneracy structures, and the CP-suppression claim — rest on setting θ₁₄ = θ₁₅ = θ₂₄ = θ₂₅ = θ₃₄ = θ₃₅ = θ_sa (§II.B). This is not merely a simplification for visualization; it is structurally load-bearing for the paper's central analytic result. Equation 43 derives the CP-dependent amplitude as proportional to c_sa = cos(θ_sa), which arises specifically because uniform active-sterile mixing rescales the entire active 3×3 submatrix by a single cosine. If the six angles differ, this clean factorization breaks: different rows of the PMNS matrix are rescaled by different cosines, and the 'universal suppression of CP-dependent amplitude by c_sa < 1' no longer holds as stated. The degeneracy valley near θ_sa ≈ θ₂₃ ≈ 0.4 (identified in §IV) is also an artifact of the universal choice — with independent angles, the cancellation condition becomes a multi-dimensional surface rather than a single point, potentially opening wider degeneracy regions. A secondary concern: the χ² in Eq. 53 fixes standard oscillation parameters to NuFit best-fit values rather than marginalizing over them. Since θ₂₃ has ~5% uncertainty and the degeneracy valley sits near θ_sa ≈ θ₂₃, allowing θ₂₃ to float would likely deepen this valley for NOvA/T2K, weakening the claimed θ_sa ≳ 0.08 bound. The DUNE result (θ_sa ≈ 0.05) is less affected since DUNE's contours lack the degeneracy valley, but it still inherits the flavor-universal assumption.","agreement_with_reader":"agree"},"referee_report":{"model":"glm-5.2","summary":"This paper studies quasi-Dirac neutrino oscillations within a 3+2 framework motivated by the inverse seesaw mechanism. Two nearly degenerate sterile neutrinos form a quasi-Dirac pair with splitting controlled by a small lepton-number-violating Majorana mass. The authors derive averaged oscillation probabilities for the appearance and disappearance channels, identifying a sterile-sector interference phase that has no analogue in non-degenerate 3+1 or 3+2 scenarios. Using GLoBES simulations of NOvA, T2K, and DUNE with realistic systematic uncertainties, they map sensitivity contours in the (theta_sa, Delta m^2_54) plane and analyse how the five-neutrino framework modifies CP-violating observables relative to the standard three-neutrino case. They find that current NOvA+T2K data constrains theta_sa >~ 0.08, while DUNE can reach theta_sa ~ 0.05 with uniform sensitivity across the full Delta m^2_54 range.","tokens_in":24208,"tokens_out":1401,"duration_ms":178919,"significance":"The paper provides a clean analytic treatment of quasi-Dirac oscillation probabilities (Eqs. 38-41) and a thorough simulation study with experimentally realistic systematics. The identification of the c_sa suppression factor governing the CP-dependent amplitude (Eq. 43) and the threshold delta_CP^thresh (Eq. 45) is a useful analytic insight. The DUNE sensitivity forecast, showing uniform coverage free of the resonance and degeneracy structures that affect NOvA and T2K, is a concrete and falsifiable prediction. The coupling of quasi-Dirac sensitivity to delta_CP measurements is an interesting and timely observation given DUNE's dual science goals. The work is well-motivated and the simulation methodology is sound.","major_comments":[{"comment":"Section II.B, Eq. (43): The claim that the CP-dependent amplitude is 'universally suppressed' by the factor c_sa < 1 is load-bearing for the paper's central analytic result, but it depends critically on the assumption that all six active-sterile mixing angles are equal (theta_14 = theta_15 = theta_24 = theta_25 = theta_34 = theta_35 = theta_sa). With independent angles, the clean factorization into a single cosine rescaling of the active 3x3 submatrix breaks down, as different rows of the PMNS matrix would be rescaled by different cosines. The authors acknowledge this in the Conclusion as a future direction, but all quantitative results — sensitivity contours, degeneracy structures, and the CP-suppression claim — rest on this simplification. The paper should either (a) explicitly state in Section II.C that Eq. (43) and the 'universal suppression' claim are derived under the flavor-univer","section":null},{"comment":"Section III.B, Eq. (53): The chi-s2 analysis fixes the standard oscillation parameters to their NuFit best-fit values rather than marginalizing over them. Given that theta_23 has approximately 5% uncertainty and the degeneracy valley identified in Section IV sits near theta_sa ~ theta_23 ~ 0.4, allowing theta_23 to float would likely deepen this valley for NOvA and T2K, potentially weakening the claimed theta_sa >~ 0.08 bound. The authors should clarify whether this effect was checked and, if not, discuss how marginalization over theta_23 (and delta_CP) would affect the quoted sensitivity limits. The DUNE result (theta_sa ~ 0.05) is less affected since its contours lack the degeneracy valley, but this should be stated explicitly.","section":null},{"comment":"Section II.B: The sterile CP phases eta_13, eta_14, and eta_24 are set to zero throughout the analysis. The phase phi_45 = (eta_14 - eta_15) - (eta_24 - eta_25) defined in Eq. (37) enters the appearance probability in Eq. (38) and could shift the sensitivity contours, particularly in the appearance channel where the sterile contribution is most pronounced. The paper should briefly discuss the expected impact of non-zero sterile CP phases on the sensitivity results, even if a full scan is deferred to future work.","section":null}],"minor_comments":[{"comment":"Table I: The range for theta_sa is listed as 0.1-1, but the sensitivity contours in Figs. 7-10 extend down to theta_sa = 0.05. Clarify whether the table range refers to the scan range or the physically motivated range.","section":null},{"comment":"Figure 2: The caption states delta_CP = 0.82 and 3.66, but the text in Section II.C discusses these values without explaining why 0.82 is chosen (it is not a standard benchmark). A brief justification would help the reader.","section":null},{"comment":"Section II.E, Eqs. (49)-(52): The decay width expressions use theta_sa as the mixing angle, but the sterile neutrino mass m_N is not explicitly defined in terms of the mass eigenvalues m_4, m_5. Clarify which mass is used (presumably the average).","section":null},{"comment":"Figure 3: The Earth-Sun and SN1987A labels appear in the legend but are not discussed in the text. Either add a brief discussion or remove them from the figure.","section":null},{"comment":"Section IV: The text states that the degeneracy valley near theta_sa ~ 0.4 arises because 'the sterile contribution to the oscillation amplitude becomes comparable to the active-active contribution.' This is a qualitative statement; a more quantitative explanation of the cancellation condition would strengthen the analysis.","section":null},{"comment":"Reference [19] (Abada et al., arXiv:2506.16390) appears to be a very recent preprint. If the present work overlaps significantly with it, the relationship should be clarified.","section":null}],"recommendation":"minor_revision","confidential_remarks":"The flavor-universal theta_sa assumption is the main vulnerability of this paper. It is not inherently wrong — it is a reasonable first scan strategy — but the paper presents results that are special-case (single-angle) as if they were general properties of the quasi-Dirac framework. The fix is primarily presentational: qualify the claims, add a brief discussion of how the results would change with independent angles, and marginalize over theta_23 in the chi-s2 if feasible. None of these require new physics or a rewrite of the core analysis. The central DUNE sensitivity forecast is likely robust since DUNE's contours lack the degeneracy structures, but the NOvA/T2K bounds may be optimistic. I lean toward minor revision because the analytic framework and simulation methodology are sound and the issues are local to the interpretation and qualification of claims."},"author_rebuttal":null,"desk_editor":{"model":"glm-5.2","letter":"The main thing to know: this paper forecasts NOvA+T2K combined sensitivity and DUNE reach for quasi-Dirac sterile neutrinos in a 3+2 inverse-seesaw framework, and derives a clean analytic result showing the 5ν appearance probability suppresses the CP-dependent amplitude by a factor c_sa = cos(θ_sa) relative to the 3ν case. The DUNE forecast (θ_sa ≈ 0.05, uniform across Δm²_54) is the strongest contribution and is credible given the simulation setup. The analytic CP-suppression argument and the identification of the atmospheric-resonance feature near Δm²_54 ~ Δm²_31 and the degeneracy valley near θ_sa ≈ θ_23 are genuinely new relative to Anamiati et al. and Abada et al. The GLoBES simulations include realistic per-experiment systematics, and the averaged oscillation probability derivations (Eqs. 38–41) are correct and follow cleanly from unitarity. The bi-event plot (Fig. 11) is honest — it shows 5ν predictions sitting inside 3ν uncertainty bands, which underscores that current experiments can't distinguish the models. The soft spot is real and the stress-test note lands it correctly: setting all six active-sterile angles equal (θ_14 = θ_15 = θ_24 = θ_25 = θ_34 = θ_35 = θ_sa) is not just a visualization choice — it's structurally load-bearing for the central analytic result. The c_sa factorization in Eq. 43 works because uniform mixing rescales the entire active submatrix by one cosine. With independent angles, different PMNS rows get different rescalings and the universal suppression claim breaks. The degeneracy valley near θ_sa ≈ 0.4 is also partly an artifact of this choice. The authors acknowledge this only in the Conclusion. A secondary issue: the χ² fixes standard parameters to NuFit best-fits rather than marginalizing over θ_23, which matters because the degeneracy valley sits right at θ_sa ≈ θ_23. Letting θ_23 float would likely deepen that valley for NOvA/T2K, weakening the claimed θ_sa ≳ 0.08 bound. The DUNE result is less affected since its contours lack the valley structure. This is a competent phenomenology paper with a real new result (the CP-suppression factor and DUNE forecast) but a load-bearing simplification that needs either justification or a robustness check. Worth a serious referee who pushes on the flavor-universality assumption and the marginalization question.","headline":"Solid quasi-Dirac sensitivity forecast for DUNE, but the flavor-universal mixing assumption is load-bearing and insufficiently stress-tested","tokens_in":24793,"tokens_out":614,"would_cite":true,"duration_ms":65225,"reading_group":"no","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"glm-5.2","headline":"DUNE can probe quasi-Dirac neutrino splitting 2× deeper than current experiments","keywords":[],"falsifier":"If DUNE data shows no deviation from 3ν predictions down to θ_sa ≈ 0.05 for any Δm²_54 in the range 10⁻⁵–10⁻¹ eV², the quasi-Dirac scenario with flavor-universal mixing at that scale is excluded. Conversely, if a deviation is found, it must be distinguishable from a non-degenerate 3+1 sterile scenario — the quasi-Dirac interference signature requires the specific cos(x54) modulation, not just a constant flux suppression.","tokens_in":23937,"feed_emoji":"🔬","tokens_out":1079,"duration_ms":159670,"temperature":0.7,"pith_summary":"Neutrinos might be neither purely Dirac nor purely Majorana but quasi-Dirac: nearly degenerate mass pairs split by a tiny amount set by lepton-number violation. This paper shows that long-baseline oscillation experiments can detect this splitting through interference effects that have no analogue in standard three-neutrino oscillations. The authors derive analytic oscillation probabilities in a five-neutrino framework where two sterile states form a quasi-Dirac pair, simulate NOvA, T2K, and DUNE using GLoBES with full systematic uncertainties, and map the two-parameter sensitivity space (sterile mixing angle θ_sa versus intra-pair splitting Δm²_54). Current NOvA+T2K data constrains θ_sa down to about 0.08; DUNE alone reaches θ_sa ≈ 0.05 across the entire splitting range. A key structural result is that the sterile mixing universally suppresses the CP-violating amplitude by a factor c_sa < 1, so quasi-Dirac sensitivity and CP-phase measurements are intrinsically coupled: DUNE's sterile search and its CP-violation program will constrain each other.","feed_headline":"","feed_subtitle":"Quasi-Dirac neutrino interference could reveal the Dirac-vs-Majorana question through oscillation physics — if DUNE sees the signature.","key_machinery":"The five-neutrino mixing matrix U = W45 R35 R34 W25 W24 R23 W15 W14 W13 R12 · DM, with all six active-sterile angles set to a single flavor-universal value θ_sa. The averaged appearance probability P(νμ→νe) = P3ν + 4|Uμ4|²|Ue4|²(1 + cos(x54)cos(x54 − φ45)), where x54 = Δm²_54 L/4E is the quasi-Dirac interference phase. The CP-suppression factor c_sa multiplying the δCP-dependent amplitude. GLoBES simulation of NOvA (810 km), T2K (295 km), and DUNE (1300 km) with Gaussian-filtered oscillation probabilities and marginalized systematic uncertainties.","core_discovery":"The paper's central result is that quasi-Dirac neutrino oscillations produce a characteristic interference signature — a slowly varying phase Δm²_54·L/4E modulating the envelope of averaged oscillations — that is qualitatively distinct from both standard three-neutrino oscillations and non-degenerate sterile scenarios. This signature is detectable at long-baseline experiments, with DUNE achieving uniform sensitivity down to θ_sa ≈ 0.05 without the degeneracy valleys and resonance artifacts that limit NOvA and T2K. The coupling between sterile mixing and CP violation means that the 5ν appearance probability is universally suppressed relative to the 3ν case by the factor c_sa, making the sign,","pith_inferences":[],"forward_implications":["DUNE's quasi-Dirac search and its CP-violation measurement are not independent programs — a joint analysis is required because sterile mixing suppresses CP-sensitive amplitudes by c_sa, biasing δCP extraction if the 5ν sector is ignored.","The atmospheric resonance near Δm²_54 ≈ Δm²_31 creates parameter degeneracies at NOvA and T2K that DUNE's longer baseline naturally washes out, making DUNE the unambiguous near-term probe.","The degeneracy valley near θ_sa ≈ θ23 ≈ 0.4, where the 5ν model can mimic 3ν spectra, represents a blind spot for current experiments that DUNE's broad-band averaging avoids.","Neutrinoless double beta decay experiments (nEXO, LEGEND) provide complementary constraints because quasi-Dirac pairs produce near-complete cancellations in the effective Majorana mass, with residuals potentially accessible at next-generation sensitivity."],"fun_headline_variants":["DUNE could catch quasi-Dirac neutrino interference invisible to NOvA and T2K","Quasi-Dirac neutrinos imprint a slow oscillation envelope DUNE can resolve","Near-degenerate sterile pair modulates neutrino oscillation averages at long baselines","DUNE sensitivity targets quasi-Dirac signature down to θ_sa ≈ 0.05 without degeneracy vall","Quasi-Dirac oscillations suppress CP appearance by c_sa — testable at DUNE"],"cache_read_input_tokens":0,"weakest_assumption_plain":"All six active-sterile mixing angles are set to a single flavor-universal value θ_sa. This collapses a fifteen-parameter sterile mixing space into one angle, and the sensitivity contours, degeneracy structures, and CP-suppression analysis all depend on this simplification. If individual angles differ substantially across flavors, the appearance and disappearance channels would be affected differently, potentially opening or closing degeneracy valleys not captured here.","fun_headline_variants_meta":{"raw":{"variants":["DUNE could catch quasi-Dirac neutrino interference invisible to NOvA and T2K","Quasi-Dirac neutrinos imprint a slow oscillation envelope DUNE can resolve","Near-degenerate sterile pair modulates neutrino oscillation averages at long baselines","DUNE sensitivity targets quasi-Dirac signature down to θ_sa ≈ 0.05 without degeneracy valleys","Quasi-Dirac oscillations suppress CP appearance by c_sa — testable at DUNE","Five-neutrino quasi-Dirac mixing reshapes CP violation at long-baseline experiments","Active-sterile splitting couples to CP violation in quasi-Dirac neutrino appearance","NOvA and T2K constrain sterile mixing; DUNE forecasts uniform quasi-Dirac sensitivity","Quasi-Dirac neutrinos leave interference distinct from standard 3ν and non-degenerate sterile","Sterile pair mass splitting Δm²_54 imprints slow phase on averaged neutrino oscillations"]},"model":"glm-5.2","effort":"low","cost_usd":0.0,"raw_usage":{"total_tokens":1075,"prompt_tokens":530,"completion_tokens":545,"prompt_tokens_details":null},"tokens_in":530,"tokens_out":545,"duration_ms":26808,"temperature":1.0,"reasoning_tokens":327,"cache_read_input_tokens":0,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-10T00:00:06.197468+00:00","model_set":{"reader":"glm-5.2"},"falsifier":"If DUNE data shows no deviation from 3ν predictions down to θ_sa ≈ 0.05 for any Δm²_54 in the range 10⁻⁵–10⁻¹ eV², the quasi-Dirac scenario with flavor-universal mixing at that scale is excluded. Conversely, if a deviation is found, it must be distinguishable from a non-degenerate 3+1 sterile scenario — the quasi-Dirac interference signature requires the specific cos(x54) modulation, not just a constant flux suppression.","supporting_citations":[],"review_version":1}