{"id":"b904b8a4-aeeb-4fa1-a09e-f7e0a2e57aa3","arxiv_id":"2507.06823","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"If the Alternative Hypothesis plus the new weak density condition holds, then asymptotically 100% of the non-trivial zeros of the Riemann zeta function are simple and on the critical line.","lead":"This paper shows that the Alternative Hypothesis, a leading rival to Montgomery's pair correlation conjecture, plus a new weak density condition, implies that almost all zeros of the Riemann zeta function are simple and lie on the critical line, with no Riemann Hypothesis assumed. The result suggests that this essential simplicity of the zeros survives regardless of which of the two rival spacing conjectures is true.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"AH2 is the decisive axiom: it selects p0=1 among AH-Pairs+AH1-compatible models, so Theorem 4's 100% conclusion is an input of the model, not a consequence of AH-Pairs.","rationale":"The reader's identification of AH2 as the weakest assumption is correct, and the stress-test sharpens it. I checked the central algebra: Theorem 2 is a valid consequence of the Gallagher-Mueller second moment and AH-Pairs, and the passage from (1.22)-(1.24) to p0=1 is sound. The load-bearing issue is not a technical error but the status of (AH2): once AH-Pairs and (AH1) are granted, (AH2) is equivalent to p0=1 up to the stated error terms, i.e., it encodes the conclusion. The (1.15) family with p0 in [1, 3/2-2/pi^2] demonstrates that AH-Pairs plus AH1 admits a continuum of density models, only one of which satisfies AH2. Thus the 100% simplicity result is a theorem about the chosen AH-Weak Density model, not about AH-Pairs. The paper is honest about this, and the conditional verdict is appropriate; I would not change the reader's CONDITIONAL verdict. No formal verification or numerical test is supplied, which supports moderate confidence.","tokens_in":10483,"tokens_out":25614,"duration_ms":276873,"concrete_test":"Construct the explicit limiting densities from (1.15) with a free parameter p0 in [1, 3/2 - 2/pi^2]: set P_j = p0 - 1/2 for j>=1 and P_{j-1/2} = 3/2 - 2/(pi^2(2j-1)^2) - p0. Verify algebraically: (i) AH1 holds for every p0; (ii) nonnegativity and (1.13) hold; (iii) the left side of (AH2) equals M(3/2 - p0) - 1/4 + O(1/M), which matches M/2 - 1/4 only when p0=1. This shows AH2 is not a consequence of AH-Pairs plus AH1, so Theorem 4 is genuinely conditional on the extra model assumption (AH2).","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim of Theorem 4 rests on the new AH-Weak Density axiom, specifically (AH2), rather than on AH-Pairs. Under AH-Pairs and (AH1), equation (1.22) gives (3/2 - P0(T)) = (1/M) sum_{j=1}^M P_{j-1/2}(T) + O(sqrt(log M)/M) + O(M(R(T)+RP(T)+1/L^2)), so (AH2) is exactly the assertion that this average is 1/2 - 1/(4M) up to errors; that forces P0(T)=1 via (1.24). Without (AH2), Corollary 2 yields only limsup P0 <= 3/2 and at least 50% simple and critical zeros. The independence of (AH2) is visible in the RH-conditional family (1.15): for each p0 in the allowed interval [1, 3/2 - 2/pi^2], the limiting densities satisfy AH-Pairs, (AH1), nonnegativity, and (1.13), but (AH2) holds only at p0=1. No theorem in the paper without (AH2) eliminates the upper-endpoint model p0=3/2 - 2/pi^2, so the 100% result is selected by the model assumption, not forced by AH-Pairs. The paper is transparent that AH-Weak Density is a model, but the headline claim should be read with that caveat clearly attached.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the implications of the Alternative Hypothesis (AH) for the simplicity and critical-line placement of zeros of the Riemann zeta-function, without assuming RH. It introduces two hypotheses: AH-Pairs, a version of AH asserting that pair spacings are close to half-integers, and a new AH-Weak Density consisting of two axioms, (AH1) and (AH2), which specify averages of the densities P_{j-1/2} and P_j. Theorem 1 shows that, under AH-Pairs, p0=1 is equivalent to Essential Simplicity. Theorem 2 is a conditional identity for the Gallagher-Mueller second-moment sum. Combining Theorem 2 with (AH1) gives Theorem 3, and then adding (AH2) yields p0=1 in Theorem 4, hence 100% simple and critical zeros. Without (AH2), Corollary 2 gives only limsup P0 ≤ 3/2 and at least 50% simple and critical zeros. The proofs are detailed and the use of known second-moment estimates is careful.","tokens_in":10871,"tokens_out":10313,"duration_ms":101164,"significance":"If AH2 were a consequence of AH-Pairs or independently well-motivated, the result would be a significant extension of the Gallagher-Mueller method beyond Montgomery's pair correlation conjecture, showing that a different pair-correlation model also forces essential simplicity. The paper is transparent about its conditional structure and correctly uses unconditional estimates of Gallagher-Mueller, Fujii, and Tsang. However, the central 100% conclusion is not a consequence of AH-Pairs alone: AH2 fixes the average of the half-integer densities and, via Theorem 3, directly forces p0=1. The RH-conditional family (1.15) shows that models with p0 in the entire allowed interval [1, 3/2 - 2/pi^2] satisfy AH-Pairs and (AH1), but AH2 holds only at p0=1. Thus the paper's main theorem is essentially an implication from an axiom chosen to encode the conclusion, rather than a demonstration that the Alternative Hypothesis itself implies essential simplicity. The 50% bounds of Corollary 2 are the strongest conclusions that do not rely on AH2.","major_comments":[{"comment":"The load-bearing step toward p0=1 is the new axiom AH2, not AH-Pairs. Indeed, substituting AH2 into (1.22) gives P0(T) = 1 + O(sqrt(log M)/M) + O(M(R(T)+RP(T)+1/L^2)), so Theorem 4's conclusion follows immediately. Conversely, under AH-Pairs and (AH1), equation (1.22) shows that p0=1 is equivalent to the averaged half-integer density tending to 1/2, which is exactly what AH2 asserts up to its stated errors. The paper even displays the RH-conditional family (1.15): for every p0 in [1, 3/2 - 2/pi^2] the limiting densities satisfy AH-Pairs, (AH1), nonnegativity, and the bound (1.13), but AH2 holds only at p0=1. Therefore the 100% result is effectively an input of the model, and without AH2 the upper-endpoint model p0=3/2 - 2/pi^2 is not eliminated. The abstract and introduction should state clearly that the 100% conclusion is conditional on the additional AH2 conjecture, and the paper should discuss what independent evidence, if any, supports AH2.","section":"Section 1, definition of AH-Weak Density"},{"comment":"(AH1) is also an additional strengthening of AH-Pairs, not a consequence of it. AH-Pairs gives errors of size O((|k|+1)R(T)) for the location of individual pair spacings, while (AH1) asserts a uniform, j-independent error O(RP(T)) for the sums P_{j-1/2}(T)+P_j(T). The footnote in the paper explains this choice heuristically, but the statement 'AH1 is obtained immediately from (1.15)' may mislead: (1.15) is a limiting formula under RH, not an error estimate under AH-Pairs. The paper should explicitly label (AH1) and (AH2) as new model assumptions distinct from AH-Pairs, and should justify or at least clearly flag the strengthening involved in making the error independent of j.","section":null}],"minor_comments":[{"comment":"The error term appears as 'O(p log M)' in the displayed equation; this should be O(sqrt(log M)) as written in (1.22).","section":"Equation (1.18)"},{"comment":"The displayed chain contains a duplicated summation symbol; the intended statement is simply that |P(T,M)| ≪ M TL, hence the sum of the densities is O(M).","section":"Equation (1.13)"},{"comment":"AH2 is stated as uniform in M on compact intervals, but the derivation of p0=1 passes to the limit M→∞ for fixed T and then lets T→∞. The sentence should be clarified so that the order of limits is explicit and no growth of M with T is required.","section":"Section 1, AH2"},{"comment":"The notation N(T, C0 R(T)) is used without re-stating that it counts pairs with |(gamma-gamma')L| ≤ C0 R(T); the symmetric version of N(T, lambda) should be defined or explicitly recalled here.","section":"Section 2, equation (2.2)"},{"comment":"There are several spacing/OCR artifacts in the title and running heads ('P AIR CORRELA TION', 'W eak', 'F aculty') that should be corrected in the final version.","section":"General formatting"}],"recommendation":"major_revision","confidential_remarks":"The paper is a careful conditional analysis, but its headline theorem rests on the new AH2 axiom, which essentially selects p0=1. I would advise the editor that the result is likely to be of interest only if the authors substantially reframe the contribution: clarify that AH-Pairs alone does not imply 100% simplicity, emphasize Corollary 2 as the strongest unconditional (within AH-Pairs) result, and discuss the independence of AH2 explicitly, perhaps by exhibiting the family of p0-values consistent with AH-Pairs and (AH1)."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Mike,\n\nYou should know about this one because it completes a program and because the headline needs a caveat. The paper shows that if you assume the Alternative Hypothesis in a sufficiently strong form, you again get 100% simple zeros on the critical line, with no RH. That part is new and the proofs look sound. But the strong form is doing real work, and the paper is honest about calling it a model.\n\nWhat is genuinely new: Theorem 1 proves that, under AH-Pairs alone, p0 = 1 is equivalent to Essential Simplicity. That is a clean bridge. Theorem 2 is a careful Gallagher-Mueller average identity, and Corollary 2 gives 50% bounds for simple and critical zeros without any new axiom. Those are solid contributions.\n\nThe soft spot is exactly where the stress-test note lands. The 100% conclusion in Theorem 4 comes from adding AH2, which asserts the average of half-integer densities is M/2 - 1/4. That assumption, combined with Theorem 3, forces p0 = 1. Without it, you only get the 50% bounds. The paper does not hide this—it explicitly says AH-Weak Density is a model—but the reader should not walk away thinking AH-Pairs alone implies simplicity. The RH-conditional family in (1.15) shows that AH-Pairs and AH1 are compatible with a range of p0 values, so AH2 is selecting the endpoint, not deriving it. That is a real limitation of the headline claim, though not a flaw in the proofs.\n\nMinor caveats: R(T) and RP(T) are assumed decreasing without rates, and the uniformity statement in AH2 is a bit unusual, but these do not affect the logic.\n\nWho is this for? Specialists in analytic number theory, particularly people working on pair correlation and the distribution of zeta zeros. It is conditional, but it is rigorous and transparent, and it moves the discussion forward. I would send it to a serious referee and recommend acceptance with a request that the authors emphasize in the introduction that Theorem 4 is a consequence of the AH-Weak Density model, not of AH-Pairs alone.\n\nBest,\n[Your name]","headline":"A careful, honest conditional paper: the 100% simplicity result is real but rests on a new model axiom (AH2) that does the decisive work; still deserves a serious referee.","tokens_in":11370,"tokens_out":1791,"would_cite":true,"duration_ms":20385,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["11M06","11M26"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper proves that the Alternative Hypothesis, the main rival to the standard pair correlation conjecture, also forces asymptotically 100% of the zeros of the Riemann zeta-function to be simple and on the critical line, with no…","keywords":["Riemann zeta-function","zeros","pair correlation","Alternative Hypothesis","simple zeros","critical line","Essential Simplicity","second moment"],"falsifier":"Compute, from the first several million zeros of $\\zeta(s)$, the densities $P_{j-1/2}(T)$ for $1\\le j\\le M$ with $M$ large, and check whether their average minus $M/2$ approaches $-1/4$ within the stated error $O(1/M)$; a systematic deviation would refute AH2 and break Theorem 4. Alternatively, a direct numerical estimate of $p_0$ differing from 1 at a scale not explained by the error terms would falsify the conclusion.","tokens_in":10257,"feed_emoji":"","tokens_out":9296,"duration_ms":88390,"temperature":0.7,"pith_summary":"This paper tries to show that a conclusion previously drawn from the standard pair correlation conjecture — that asymptotically 100% of the nontrivial zeros of the Riemann zeta-function are simple and lie on the critical line — is not an accident of that one conjecture. Under the Alternative Hypothesis, the main competing picture of how zeros are spaced, it formulates a precise 'weak density' version and proves the same 100% conclusion, without assuming the Riemann Hypothesis. The load-bearing step is a weighted average identity (Theorem 2) that links the pair densities $P_{k/2}(T)$ to the second moment of zeros in short intervals; adding a new half-integer density sum rule forces $p_0 = 1$, which is equivalent to Essential Simplicity. If correct, the simplicity question is settled under either competing conjecture and the Riemann Hypothesis is not required.","feed_headline":"The Alternative Hypothesis also yields 100% simple zeta zeros","feed_subtitle":"Both competing pair-correlation conjectures force the same 100% simplicity conclusion.","key_machinery":"The central objects are the pair densities $P_{k/2}(T)$, defined as the number of pairs of zeros $0<\\gamma,\\gamma'\\le T$ whose normalized distance $(\\gamma-\\gamma')L$ lies in $(k/2-1/4,\\, k/2+1/4]$, divided by $TL$. The argument runs on a weighted identity (Theorem 2) obtained by equating an unconditional second-moment asymptotic for zeros in short intervals with a combinatorial expansion of the same weighted count under AH-Pairs. The new auxiliary input AH-Weak Density (conditions AH1 and AH2) then turns this identity into asymptotic formulas for $P_0(T)$, yielding $p_0=1$. The half-integer sum rule AH2 is the specific new mechanism that fixes the average of the half-integer densities to $M/2 - 1/4$.","core_discovery":"The paper's central claim is that, under the Alternative Hypothesis for pairs of zeros (AH-Pairs) together with the newly stated Alternative Hypothesis for Weak Density (AH-Weak Density), the limiting density $p_0$ of pairs of zeros within a quarter of the average spacing equals 1. By Theorem 1, $p_0 = 1$ is equivalent to Essential Simplicity, and Essential Simplicity is already known to imply that asymptotically 100% of zeros are simple and on the critical line. The proof does not invoke the Riemann Hypothesis at any point; RH is used only to motivate the formulation. The auxiliary hypothesis splits into (AH1), a fixed asymptotic value for the sum of adjacent half-integer and integer pair densities, and (AH2), a half-integer density sum rule giving average $M/2 - 1/4$. The latter is the only new input needed to pass from the average relations of Theorem 3 to $p_0=1$.","pith_inferences":["We would add that AH2 is the real new hypothesis: it is not derived from AH-Pairs, and the paper's 100% conclusion collapses to the weaker 50% bounds if AH2 fails, so the credibility of the theorem rests on this specific average rule.","Going beyond the paper, the same weighted second-moment technique could be applied to other $L$-functions or to the derivative of $\\zeta$, where the Alternative Hypothesis has also been studied; one would expect an analogous half-integer density sum rule to control the proportion of simple zeros there.","We infer that if the 100% conclusion holds under both the standard pair correlation conjecture and the Alternative Hypothesis, the simplicity and on-line properties of zeta zeros may be insensitive to the fine details of vertical pair correlation, pointing toward a model-independent theorem that would hold under any pair-correlation conjecture of the same shape.","Numerically, one can test AH2 at finite $T$ with the first few million zeros: the prediction that the average of the first $M$ half-integer densities is $M/2 - 1/4$ is precise enough to be distinguished from the standard pair-correlation prediction, which behaves differently."],"forward_implications":["Under AH-Pairs and AH-Weak Density, asymptotically 100% of the zeros of $\\zeta(s)$ are simple and lie on the critical line, with no Riemann Hypothesis assumption (Theorem 4).","Assuming AH-Pairs, $p_0=1$ and Essential Simplicity are equivalent (Theorem 1), so the 100% conclusion is interchangeable with the pair-repulsion and simplicity statement (ES).","If only AH1 is assumed, the weaker conclusion holds: at least 50% of zeros are simple and at least 50% are on the critical line, with $\\limsup_{T\\to\\infty} P_0(T) \\le 3/2$ (Corollary 2).","Under RH, $p_0=1$ together with the known density relations determines all limiting densities $p_{k/2}$: $1/2$ for nonzero even $k/2$ and $1/2 - 2/(\\pi^2 k^2)$ for odd $k/2$ (Equation (1.17))."],"supporting_citations":[{"why":"The companion paper that established the 100% conclusion under the standard pair correlation conjecture and supplies the second-moment method for zeros in short intervals used here.","marker":"[GLSS25]"},{"why":"Provides the second-moment method for zeros in short intervals, packaged here as Proposition 1, that links the weighted pair count to the square integral of the counting function.","marker":"[GM78]"},{"why":"Gives the unconditional bound on the number of pairs of zeros in short intervals (Lemma 9) that keeps the densities $P_{k/2}(T)$ bounded and justifies the error estimates.","marker":"[GM87]"},{"why":"Formulates AH-Pairs and, under RH, derives the density relations that serve as the model for AH-Weak Density.","marker":"[BGSTB25a]"},{"why":"Introduced the standard pair correlation conjecture and the original method that the paper parallels; its RH-conditional formulas motivate (1.14)-(1.17).","marker":"[Mon73]"},{"why":"Provides the standard zero-counting estimate $N(T) \\sim TL$ and the Riemann-von Mangoldt formula used in the second-moment estimates.","marker":"[Tit86]"},{"why":"Supplies part of the variance estimate for $S(t)$ in Proposition 2, needed for the $\\frac{1}{\\pi^2}\\log(2+\\lambda)$ term in the second-moment identity.","marker":"[Fuj74]"},{"why":"Extends the variance estimate for $S(t)$, also used in Proposition 2.","marker":"[Fuj81]"},{"why":"Provides an additional variance estimate for $S(t)$ used in Proposition 2.","marker":"[Tsa84]"}],"fun_headline_variants":["Both pair correlation theories force 100% simple zeros","Alternative Hypothesis also proves all zeta zeros simple","New hypothesis yields 100% simple zeros without RH","No RH needed: AH implies 100% simplicity"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The paper's conclusion depends on the auxiliary conjecture AH2, which asserts that the average of the first $M$ half-integer pair densities tends to $M/2 - 1/4$; this is not derived from AH-Pairs and is the new input that forces $p_0 = 1$.","fun_headline_variants_meta":{"raw":{"variants":["Both pair correlation theories force 100% simple zeros","Alternative Hypothesis also proves all zeta zeros simple","New hypothesis yields 100% simple zeros without RH","No RH needed: AH implies 100% simplicity"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000529,"raw_usage":{"total_tokens":2500,"prompt_tokens":847,"completion_tokens":1653,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":463,"completion_tokens_details":{"reasoning_tokens":1591}},"tokens_in":463,"tokens_out":1653,"duration_ms":14194,"temperature":1.0,"reasoning_tokens":1591,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T18:54:05.566894+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute, from the first several million zeros of $\\zeta(s)$, the densities $P_{j-1/2}(T)$ for $1\\le j\\le M$ with $M$ large, and check whether their average minus $M/2$ approaches $-1/4$ within the stated error $O(1/M)$; a systematic deviation would refute AH2 and break Theorem 4. Alternatively, a direct numerical estimate of $p_0$ differing from 1 at a scale not explained by the error terms would falsify the conclusion.","supporting_citations":[],"review_version":1}