{"id":"97bd76cb-18a7-499e-85b2-65d36b8416f0","arxiv_id":"2411.17327","paper_version":1,"verdict":"REJECT","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"high","formal_verification":"none","parameter_count":1,"one_line_summary":"The paper claims exact infinite-series representations for arithmetic sums over solutions of kb^2 plus or minus da^2 equals N, and an equivalent reformulation of the Riemann hypothesis based on them.","lead":"This paper derives very long infinite-series formulas for sums over integer solutions of quadratic Diophantine equations, and rewrites the Robin-Lagarias criterion for the Riemann hypothesis using those formulas. The identities are not verified numerically, and one printed term appears inconsistent with an earlier equation, so the central formulas are not currently supported.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Proposition 18 does not follow from Eq. (4.7): substituting q1(R)/R^2 and multiplying by R^{5/2} gives a first term π^2 R^{3/2}/[3(e^{2πt}-1)], not π^2 R^{-1/2}/[3(e^{2πt}-1)]. The RH inequality inherits this factor-R^2 error.","rationale":"The reader's weakest_assumption correctly targets the only place where the central claim can fail internally: the passage from the q1-series identity (4.7) to the σ(N) series (6.2). My independent check of the exponents confirms the substitution is not algebraically exact as printed. In fact the printed first exponent R^{-1/2} would require a denominator R^3 in (4.7), while the printed (4.7) has denominator R; the discrepancy is a factor R^2. This is more severe than a mere typo in one term because the same erroneous first term appears in the RH inequality (1.13) and (6.6), so Proposition 3 is unsupported. I do not see an independent reason to reject the inversion framework; the lemmas on convergence are elaborate and the defect is localized. But the submitted text contains no proof of the corrected exponent and no numerical verification, so the central claim cannot be accepted as written. Hence the reader's REJECT verdict should stand.","tokens_in":33826,"tokens_out":11983,"duration_ms":109771,"concrete_test":"Symbolically redo the §6 substitution: in Eq. (6.1), replace q1(R)/R^2 by the RHS of Eq. (4.7) with k=1, N→4N, c=a^2, then multiply by R^{5/2}; verify the coefficient of π^2/(3(e^{2πt}-1)) is R^{3/2}, not R^{-1/2}. A numerical spot check: for N=2, t=1, evaluate the printed RHS of Proposition 18 with the r,m-sums truncated at 50 (the series converge exponentially) and compare with σ(2)=3; the first-term exponent error alone changes the value by about π^2/(3(e^{2π}-1))·(27 − 1/3) ≈ 0.042. The exact symbolic check is definitive and does not depend on truncation.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing step is the substitution of Eq. (4.7) into Eq. (6.1) that produces Proposition 18. With R = 4N+a^2, k = 1, c = a^2, and N in (4.7) replaced by 4N, the simplified expression for q1(R)/R^2 begins with π^2/[3R(e^{2πt}-1)]. Multiplying by the factor R^{5/2} in (6.1) gives π^2 R^{3/2}/[3(e^{2πt}-1)]. The printed Proposition 18 instead has R^{-1/2}/[3(e^{2πt}-1)], discrepant by R^2 = (4N+a^2)^2. Equivalently, matching the printed exponent would require the first term of (4.7) to have denominator R^3, not R. Since all other displayed terms in (6.2) have the exponents predicted from (4.7), this is a local algebra error in the leading term, not a reindexing. Consequently the asserted identity σ(N) = ... is false as printed, and inequality (1.13)/(6.6) is not a valid equivalent form of Lagarias' criterion. The paper supplies no numerical check that would have caught this.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a general method, based on a Fourier-inversion lemma for series of partial fractions, to derive analytic expressions for sums over positive integer solutions of Diophantine equations of the forms k b^2 + d a^2 = N and k b^2 - d a^2 = N. It applies the method to the arithmetic functions q_{k,s}(N), obtains explicit infinite-series formulas for certain divisor-type sums, and finally claims a new equivalent form of the Robin-Lagarias criterion for the Riemann hypothesis by writing sigma(N) as an infinite series with a free parameter t.","tokens_in":34147,"tokens_out":21851,"duration_ms":197394,"significance":"If the main identities were correct, the paper would provide striking explicit analytic representations for arithmetic sums that are normally accessible only through modular or sieve-theoretic methods, and the parametrized reformulation of the Lagarias criterion would be a new equivalent statement of the Riemann hypothesis. The method itself, particularly the inversion theorem in Section 2, is potentially of independent interest. However, the central sigma(N) identity contains a concrete algebra error in the leading term, so the main claim is not established. The Riemann hypothesis part is, in any case, a direct restatement of Lagarias' theorem once sigma(N) is identified with the proposed series; it supplies no independent evidence about the Riemann hypothesis. The paper also provides no numerical verification of any of its long identities, which is a serious gap in view of the complexity of the formulas.","major_comments":[{"comment":"Proposition 18 does not follow from the stated substitution. In Eq. (6.1), the term for sigma(N) is q1(4N+a^2)/(4N+a^2)^2 multiplied by (4N+a^2)^{5/2}. Substituting Eq. (4.7) with k=1, c=a^2, and N replaced by 4N gives, for R=4N+a^2, the leading contribution pi^2/(3R(e^{2 pi t}-1)) times R^{5/2}, i.e. pi^2 R^{3/2}/(3(e^{2 pi t}-1)). The first term printed in Eq. (6.2) is instead pi^2/(3(e^{2 pi t}-1) R^{1/2}), which differs from the substituted value by a factor R^2 = (4N+a^2)^2. Thus Eq. (6.2) is not the result of substituting Eq. (4.7) into Eq. (6.1), and Proposition 18 is unproved as stated.","section":"Section 6, Eqs. (4.7), (6.1), (6.2)"},{"comment":"Since the inequality in Proposition 3, Eq. (1.13), is exactly the left-hand side of Eq. (6.6), and since Eq. (6.6) is obtained from the incorrect Proposition 18, the claimed equivalence with the Riemann hypothesis is not supported. The paper does not establish that the left side of (1.13) equals sigma(N), so the parametrized inequality is not a proved equivalent of Lagarias' criterion. This is a load-bearing failure: the central advertised consequence of the paper depends on the invalid substitution.","section":"Section 6, Proposition 3 and Eq. (6.6)"}],"minor_comments":[{"comment":"There are numerous typographical and OCR-style defects in the long displayed formulas, such as unbalanced parentheses in the definition of G_{M,t,k} in Eq. (1.8) and inconsistent exponents in several places; these make independent verification substantially harder.","section":"Throughout"},{"comment":"The paper describes Proposition 3 as a 'possible improvement' of the Robin-Lagarias criteria, but the free parameter t enters only through an identity that is claimed to hold for all t>0; without the sigma(N) identity the statement is merely a restatement of Lagarias' theorem, and with it it is an equivalent reformulation rather than an improvement.","section":"Section 1 and Section 6"},{"comment":"The star convention for singular summands is introduced informally and used in the derivation of Propositions 14-16, but the equivalence between the original finite sum and the star-modified infinite series is not proved. This is not directly involved in Proposition 18, but it is a gap in the earlier derivations.","section":"Section 5, Eqs. (5.11)-(5.13)"},{"comment":"The closed forms in Lemma 12 are stated after only a one-line indication of proof; since later substitutions in Sections 5 and 6 depend on exact constants, a fuller derivation or numerical cross-check for these constants would be needed.","section":"Section 4, Lemma 12"}],"recommendation":"reject","confidential_remarks":"The paper contains a central algebraic error in the derivation of the sigma(N) series, and the advertised Riemann hypothesis criterion inherits that error. The absence of any numerical verification is particularly unfortunate, since a single evaluation at small N and t would have exposed the discrepancy. I do not see a way to repair the manuscript within its current scope without redoing the core computation and rechecking the long formulas. Reject."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know. The inversion technique in Sections 2–3 is a genuine idea, and the q_{k,s} representation appears new relative to the cited literature. But the flagship application—the σ(N) formula and the Riemann-hypothesis criterion that rides on it—is wrong as printed.\n\nThe stress-test check holds up. Putting R = 4N+a^2 and substituting (4.7) into (6.1), the first term of (4.7) contributes π^2 R^{3/2}/[3(e^{2πt}-1)] after multiplying by R^{5/2}, not the π^2 R^{-1/2}/[3(e^{2πt}-1)] that appears in Proposition 18. Every other term in (6.2) has the exponent predicted from (4.7), so this is a local but load-bearing algebra error, not a reindexing. Because the leading factor is R^2 off, fixing the exponent would make the expression far too large to equal σ(N); this is not a one-character typo. Proposition 3's inequality (6.6) therefore does not follow from Lagarias's criterion.\n\nThe paper does real work before that point. Lemma 4 and Theorem 5 give a plausible inversion of partial-fraction series, and the convergence proofs in Lemmas 7, 10, and 13 are detailed and mostly convincing. The derivation is not circular and no fitted constants appear. What is missing is any numerical check; one evaluation for small N would have caught the R^2 discrepancy. The star convention for singular terms is stated but not justified—taking singular terms as zero must be shown to preserve equality as a limit. And as the reader notes, the RH 'improvement' is a definitional restatement of Lagarias once you subtract the false identity; it contributes no independent evidence.\n\nWho gets value: someone interested in explicit series for representation numbers by quadratic forms could mine the method, but they should treat the convergence and the singular-term handling with care. I would not accept this version, but I would send it to a serious referee rather than desk reject it; the inversion idea is substantive enough to warrant a careful check and perhaps a salvage of the σ(N) application.\n\nRecommendation: engage with it as a borderline revise-and-resubmit, if the journal has the patience for very long formulas.","headline":"The inversion method is worth a look, but the sigma(N) formula and the RH criterion collapse on a factor-R^2 algebra error.","tokens_in":34615,"tokens_out":4765,"would_cite":false,"duration_ms":40844,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["11A25","11D09","11M26"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper claims that the sum-of-divisors function has an explicit infinite-series representation, which converts the Riemann hypothesis into a concrete inequality holding for every real parameter $t>0$.","keywords":["analytic representation","arithmetic functions","Diophantine equations","sum of divisors","Riemann hypothesis","Robin-Lagarias criterion","Mittag-Leffler expansion","partial fraction inversion"],"falsifier":"Evaluate the right-hand side of Proposition 18 for a small integer such as $N=2$ at a fixed real $t>0$, truncating the convergent series, and compare with $\\sigma(2)=3$; if the value does not approach $\\sigma(2)$ as more terms are included, the claimed representation fails. Alternatively, check the denominator exponent in the first term of Eq. (4.7): if it is genuinely $(N+c)$, then Propositions 18 and 3 cannot follow from the stated substitution.","tokens_in":100,"feed_emoji":"∑","tokens_out":4262,"duration_ms":98110,"temperature":0.7,"pith_summary":"The paper develops a method for writing certain arithmetical sums as convergent infinite series, by inverting a formal partial-fraction series through a Mittag-Leffler expansion. It applies the method to sums over positive integer solutions of $kb^2+da^2=N$ and $kb^2-da^2=N$, and then to the sum-of-divisors function $\\sigma(N)$. The central consequence is Proposition 3: the Riemann hypothesis is equivalent to the inequality (1.13) holding for all $N>1$ for any fixed real $t>0$, because the left side is claimed to equal $\\sigma(N)$. If the representation is exact, the Lagarias criterion becomes a one-parameter family of elementary-looking inequalities. A sympathetic reader should care because this would be a new analytic handle on a classical arithmetic function and on the Riemann hypothesis.","feed_headline":"RH shown equivalent to a single explicit inequality","feed_subtitle":"If the derivation holds, sigma(N) gets an infinite-series formula that makes the Lagarias criterion a concrete, checkable bound.","key_machinery":"The engine is an inversion theorem (Theorem 5) for the formal series $\\sum_{n\\ge 1} f(n)/(n+z)$: if this series equals $F(z)$, then $f(N)$ is recovered as an infinite series of differences $F(k-N\\pm it)-F(k-N\\mp it)$ weighted by integrals $\\int_0^1 \\cos(\\pi k\\beta)/\\cosh(\\pi\\beta t)\\,d\\beta$. Applied to $F(z)$ from Lemma 6, this yields explicit representations for the square-indicator functions $q_{k,s}(N)$, hence, after summing over $a$, for the divisor sums in Propositions 15 and 16, and finally for $\\sigma(N)$ in Proposition 18.","core_discovery":"The paper's central claim is that the arithmetic function $\\sigma(N)$ admits the explicit series representation (6.2), obtained by substituting the analytic expression for $q_1(4N+a^2)/(4N+a^2)^2$ into the divisor-sum identity $\\sigma(N)=q_1(N)\\sqrt{N}+\\sum_{a=1}^{N-1} q_1(4N+a^2)\\sqrt{4N+a^2}$. Because the left side of (6.2) is asserted to equal $\\sigma(N)$ for every real $t>0$, the Robin-Lagarias inequality $\\sigma(N)<H_N+e^{H_N}\\log H_N$ is claimed equivalent to the inequality (1.13) for any fixed $t>0$. The paper thereby presents proving or disproving the Riemann hypothesis as proving or disproving a single concrete inequality that does not refer to the zeta function.","pith_inferences":["The free parameter $t$ creates a family of equivalent formulations of the Riemann hypothesis; one could try to choose $t$ to make numerical verification of the inequality easier.","If the series representation for $\\sigma(N)$ is genuinely convergent and uniform in $N$, it might connect divisor sums to special values of hyperbolic and trigonometric sums, opening new comparisons with known estimates.","The denominator mismatch noted in the weakest-assumption field is directly testable by machine computation before any serious appeal to the Riemann hypothesis is made."],"forward_implications":["If Proposition 18 holds, $\\sigma(N)$ is given by a convergent infinite series involving hyperbolic functions and the auxiliary quantities $G_{N,t}$.","The Riemann hypothesis becomes equivalent to the single inequality (1.13) for any fixed real $t>0$.","The Robin and Lagarias bounds become accessible through term-by-term estimates of the series representation.","The method extends to sums with $b^{4s}$ and more general expressions $y(a)$ in the Diophantine equation, as the paper remarks after Proposition 16."],"supporting_citations":[{"why":"Supplies the Mittag-Leffler expansion formula (1.2) from Ramanujan's notebook that underlies the series evaluations throughout the paper.","marker":"[1]"},{"why":"States the Lagarias criterion, the elementary inequality that Proposition 3 rewrites using the paper's series representation of $\\sigma(N)$.","marker":"[2]"},{"why":"Provides the Fourier-coefficient recovery and the Weierstrass M-test used in Theorem 5 and Lemma 7.","marker":"[4]"},{"why":"Defines the Riemann hypothesis whose criterion is being targeted.","marker":"[5]"},{"why":"Gives Robin's criterion for the Riemann hypothesis, the precursor that Lagarias's inequality sharpened.","marker":"[6]"}],"fun_headline_variants":["New series for sigma(N) gives direct RH inequality","Analytic sums of divisor functions target Riemann hypothesis","Explicit formula for sigma(N) could prune Robin-Lagarias proof","Diophantine identities reduce RH to one explicit bound","Sigma(N) series hints at simplified RH verification"],"cache_read_input_tokens":36736,"weakest_assumption_plain":"The central claim depends on the substitution step from Eq. (4.7) into identity (6.1) being algebraically exact, and the printed equations show a denominator mismatch ($N+c$ versus $(N+c)^2$) in the first term that would break that exactness if it is not a typographical error.","fun_headline_variants_meta":{"raw":{"variants":["New series for sigma(N) gives direct RH inequality","Analytic sums of divisor functions target Riemann hypothesis","Explicit formula for sigma(N) could prune Robin-Lagarias proof","Diophantine identities reduce RH to one explicit bound","Sigma(N) series hints at simplified RH verification"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00017,"raw_usage":{"total_tokens":1206,"prompt_tokens":818,"completion_tokens":388,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":434,"completion_tokens_details":{"reasoning_tokens":310}},"tokens_in":434,"tokens_out":388,"duration_ms":4237,"temperature":1.0,"reasoning_tokens":310,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T12:15:35.027293+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Evaluate the right-hand side of Proposition 18 for a small integer such as $N=2$ at a fixed real $t>0$, truncating the convergent series, and compare with $\\sigma(2)=3$; if the value does not approach $\\sigma(2)$ as more terms are included, the claimed representation fails. Alternatively, check the denominator exponent in the first term of Eq. (4.7): if it is genuinely $(N+c)$, then Propositions 18 and 3 cannot follow from the stated substitution.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the Mittag-Leffler expansion formula (1.2) from Ramanujan's notebook that underlies the series evaluations throughout the paper."},{"cited_title":"An Elementary Problem Equivalent t o the Riemann Hypothesis","cited_arxiv_id":null,"evidence_quote":"States the Lagarias criterion, the elementary inequality that Proposition 3 rewrites using the paper's series representation of $\\sigma(N)$."},{"cited_title":"Ueber die Anzahl der Primzahlen unter e iner gegebenen Gr¨ osse","cited_arxiv_id":null,"evidence_quote":"Defines the Riemann hypothesis whose criterion is being targeted."},{"cited_title":"Grandes valeurs de la fonction somme des divis eurs et hypoth` ese de Riemann","cited_arxiv_id":null,"evidence_quote":"Gives Robin's criterion for the Riemann hypothesis, the precursor that Lagarias's inequality sharpened."}],"review_version":1}