{"id":"f9d24438-79e7-4922-8aa8-43e62d3e5733","arxiv_id":"2606.10331","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"When both n and n+12 are products of two distinct primes, the jump-12 equations force n=p(p+8) with (p,p+2,p+6,p+8) prime; prime powers never work.","lead":"Using the definitions of two classic number-functions, this paper proves that the only semiprime values of n where both functions jump by exactly 12 when n becomes n+12 are the products p(p+8) built from prime quadruplets, and that no prime power can work. It also checks every n up to one trillion and finds no exceptions to the broader conjecture.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 3.1's proof has a reversed inequality in Subcase 3.2: φ(N)/N is ≥, not ≤, the displayed product, so the contradiction does not follow. The no-prime-power theorem is unproved as written.","rationale":"The reader's weakest assumption focused on computational reproducibility, but the more load-bearing concern is a genuine mathematical error in the proof of Theorem 3.1, which the abstract explicitly claims ('we prove that no solution n can be a single prime power'). The reversed inequality in Subcase 3.2 invalidates the stated contradiction, so the theorem is unsupported. This is not a minor typo: the argument's conclusion depends on the false direction. The computational search, even if perfectly reproducible, cannot substitute for the proof. The paper's other results (Theorem 2.2, Corollary 2.4) appear sound, but the central contribution is incomplete. I therefore recommend rejecting the paper in its current form; a corrected proof of Theorem 3.1 would be required for acceptance.","tokens_in":8984,"tokens_out":16068,"duration_ms":143546,"concrete_test":"Test the inequality numerically and re-derive the subcase. Take p=5, r1=7, r2=11, r3=13, N=7·11·13=1001: φ(N)/N = 720/1001 ≈ 0.719, while (1−1/7)(1−1/9)(1−1/13) = (6/7)(8/9)(12/13) ≈ 0.703. The claimed '≤' fails. Independently re-prove Subcase 3.2 or search for a counterexample to Theorem 3.1 (e.g., exact φ(p^k+12) − φ(p^k) for p<10^4, 2≤k≤10^2); if no counterexample emerges, the theorem may be true but the manuscript's proof remains invalid.","verdict_should_be":"REJECT","load_bearing_attack":"Section 3, Subcase 3.2 contains a mathematically false inequality. After establishing φ(N)/N > (p−1)/p, the authors write:\n\nφ(N)/N ≤ (1−1/(p+2))(1−1/(p+4))(1−1/r3).\n\nBut because r1≥p+2 and r2≥p+4, we have 1−1/r1 ≥ 1−1/(p+2) and 1−1/r2 ≥ 1−1/(p+4). Including the remaining factors (<1) gives φ(N)/N ≥ (1−1/(p+2))(1−1/(p+4))(1−1/r3). The displayed '≤' is therefore reversed. With the correct '≥', the chain 'φ(N)/N > (p−1)/p ⇒ product > (p−1)/p' no longer follows, and the contradiction (1/r3 < 0) evaporates. Consequently Theorem 3.1 — a central claim of the paper — is not proved as written. The earlier k=2 line 'p+1−(p+2) ≥ p+4' is also algebraically incoherent, though that contradiction is recoverable. The computational search does not repair the proof gap.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies composite integers n satisfying the simultaneous jump conditions φ(n+12)=φ(n)+12 and σ(n+12)=σ(n)+12. The main results are: (i) Theorem 2.2, a complete characterization when both n and n+12 are squarefree semiprimes, namely n=p(p+8) with (p,p+2,p+6,p+8) a prime quadruplet; (ii) Theorem 3.1, asserting that no prime power p^k satisfies even the φ-jump alone; and (iii) a computational search up to 10^12 reporting exactly 166 solutions, all of the predicted semiprime form and all congruent to 65 mod 72. The paper is motivated by Conjecture 1.1, attributed to Stephan and McCranie, and notes that a proof of Conjecture 1.1 together with infinitely many solutions would imply infinitely many prime quadruplets.","tokens_in":9331,"tokens_out":17974,"duration_ms":152963,"significance":"If the results stand, the paper makes a clean contribution to an old jump-condition problem: Theorem 2.2 is an elegant, fully self-contained proof that within the semiprime class the prime-quadruplet construction is necessary and sufficient. The computational extension of the OEIS data by three orders of magnitude is also useful evidence for Conjecture 1.1. However, the proof of Theorem 3.1 contains a reversed inequality in Subcase 3.2, so the non-existence of prime-power solutions is not established as written. The value of the paper therefore depends on a corrected proof of that theorem; the computational search does not repair the logical gap.","major_comments":[{"comment":"The displayed inequality φ(N)/N ≤ (1−1/(p+2))(1−1/(p+4))(1−1/r3) has the wrong direction. Since r1≥p+2 and r2≥p+4, we have 1−1/r1 ≥ 1−1/(p+2) and 1−1/r2 ≥ 1−1/(p+4); the remaining factors are also at least 1−1/r3. Thus φ(N)/N is ≥, not ≤, that product (and is larger still when there are more than three prime factors). Consequently the later chain leading to 1/r3<0 is invalid. The correct inequality is compatible with φ(N)/N>(p−1)/p without contradiction. Theorem 3.1 is therefore not proved as written, and the computational search in Section 4 does not supply the missing argument.","section":"§3, Subcase 3.2"},{"comment":"The line 'p+1−(p+2) ≥ p+4' is not a valid substitution. From r2 = p+1−r1 and r1≥p+2 one obtains r2≤−1, which already contradicts r2≥p+4. The intended contradiction is recoverable, but the written derivation is algebraically wrong and should be corrected.","section":"§3, Subcase 3.1, k=2"},{"comment":"The claim of a 'complete search up to 10^12' is not independently verifiable from the manuscript. The cited GitHub repository [6] is not pinned to a commit, and no build/run instructions, compiler flags, or checksums for the 166-solution table are given. Since the abstract's empirical claim depends on the correctness of the custom C++ segmented sieve, the authors should provide a permanent artifact (DOI/commit) and enough detail for the computation to be reproduced.","section":"§4 and Abstract"}],"minor_comments":[{"comment":"The abstract has '10 12' and §4 has '1012'; these should be typeset as 10^12.","section":"Abstract and §4"},{"comment":"Reference [6] is a GitHub repository without a version, commit hash, or access date; a permanent identifier and retrieval date should be given.","section":"References"},{"comment":"The phrase 'no solutions can be composed of a single prime power' could be clarified: Theorem 3.1 actually shows that no prime power satisfies the φ-jump alone, which is stronger than needed for the simultaneous conditions.","section":"Abstract"}],"recommendation":"major_revision","confidential_remarks":"The reversed inequality in Subcase 3.2 is a genuine mathematical error, not a merely typographical issue; the direction of the inequality follows immediately from the established lower bounds on r1 and r2. I have not determined whether Theorem 3.1 can be salvaged by a different argument. The rest of the paper, especially Theorem 2.2, is sound and well presented. A revision with a correct proof of Theorem 3.1 and a properly pinned computational artifact would be appropriate for the journal."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The semiprime theorem is the real contribution. Theorem 2.2 reduces both jump conditions to p+q=p'+q', then to a difference-of-squares equation u^2-v^2=48, and the case analysis is clean. It settles the squarefree semiprime case of McCranie's conjecture, and the corollary about n≡65 mod 72 is a nice bonus. If the paper were only this, I'd be happy with it.\n\nThe prime-power theorem has a serious gap. In Subcase 3.2, after establishing φ(N)/N > (p−1)/p, the authors write\n\n  φ(N)/N ≤ (1−1/(p+2))(1−1/(p+4))(1−1/r3).\n\nThe inequality is backwards. Since r1 ≥ p+2 and r2 ≥ p+4, and 1−1/r is increasing in r, the product of the actual factors is ≥ the displayed product, not ≤. The stress-test note is correct. With the correct inequality, the chain leading to 1/r3 < 0 collapses, and the contradiction evaporates. Also, if N has more than three distinct prime factors, the extra factors make φ(N)/N even smaller than the three-factor product, so the displayed upper bound is not true in that case either. Theorem 3.1 is therefore not proved as written. The k=2 line in Subcase 3.1 is also sloppy, though that part looks recoverable.\n\nThe computational extension to 10^12 with 166 solutions is plausible and genuinely useful evidence for the conjecture. But the code is a custom C++ segmented sieve with no commit hash, and the appendix table is not machine-readable, so the computation is not independently reproducible from the paper. That is a presentation weakness, not a mathematical claim.\n\nNet: the paper is worth engaging with. Theorem 2.2 alone is a solid partial proof of McCranie's characterization, and the prime-power statement may well be true — the gap is a local inequality mistake, not an obvious fatal flaw. But the abstract's claim about prime powers should not stand in the current version. A serious referee should ask for a corrected proof of Subcase 3.2 or a withdrawal of that claim until fixed.","headline":"Theorem 2.2 (the semiprime characterization) is correct and worth citing; Theorem 3.1 (no prime powers) has a reversed inequality in Subcase 3.2 and is not proved as written.","tokens_in":9774,"tokens_out":4338,"would_cite":true,"duration_ms":44097,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["11N64","11N05","11Y70"],"pacs":[],"model":"deepseek-v4-flash","headline":"For products of two distinct primes, the simultaneous 12-jump in φ and σ happens exactly for prime quadruplets.","keywords":["arithmetic functions","Euler's totient function","sum-of-divisors function","prime quadruplets","semiprimes","jump conditions","computational search","prime k-tuple conjecture"],"falsifier":"Run an independent implementation of the same segmented sieve over n∈[2,10^12] and check that it returns exactly the same 166 solutions; any discrepancy would falsify the claimed completeness. Alternatively, a single composite n not of the form p(p+8) with (p,p+2,p+6,p+8) prime and satisfying both jump conditions would falsify the open conjecture, and a single prime power p^k satisfying φ(p^k+12)=φ(p^k)+12 would falsify Theorem 3.1.","tokens_in":8911,"feed_emoji":"🧮","tokens_out":8220,"duration_ms":67353,"temperature":0.7,"pith_summary":"This paper tries to pin down when a composite integer n makes both Euler's totient φ and the sum-of-divisors σ jump by exactly 12 when n moves to n+12. It proves that if n and n+12 are both products of two distinct primes, the simultaneous jumps occur exactly when n=p(p+8) with (p,p+2,p+6,p+8) a prime quadruplet. It also proves that no prime power n can satisfy the jumps. A complete computer search up to 10^12 turned up exactly 166 solutions, all of this prime-quadruplet form and all congruent to 65 modulo 72. If the full characterization is true and infinitely many such n exist, then infinitely many prime quadruplets exist.","feed_headline":"Semiprime solutions to the 12-jump are exactly prime quadruplets","feed_subtitle":"All 166 solutions up to 10^12 fit the pattern; proof covers semiprimes and rules out prime powers.","key_machinery":"The central identity is u^2-v^2=48, where u=q-p and v=q'-p' are the prime gaps inside n and n+12. Because u and v share parity, the factorization 48=(u-v)(u+v) yields a finite list of candidates, and the only survivor is (u,v)=(8,4). That single pair forces q=p+8 and q'=p+4, and combined with p+q=p'+q' reproduces the quadruplet. For the prime-power exclusion, the machinery is the elementary bound φ(m)≤m-1 plus the observation that any repeated prime factor of n+12 must also divide n+12-φ(n+12), which contradicts the reduced equation.","core_discovery":"The main result, Theorem 2.2, states that if n=pq and n+12=p'q' are both products of two distinct primes with p<q and p'<q', then the two jump conditions φ(n+12)-φ(n)=12 and σ(n+12)-σ(n)=12 hold together if and only if p≥5, q=p+8, and {p',q'}={p+2,p+6}. The proof reduces both equations to the equality p+q=p'+q', then uses the discriminant relation (q-p)^2-(q'-p')^2=48. Factoring 48 gives only three possible pairs of square gaps; parity rules out all but (8,4), forcing the prime-quadruplet spacing. A separate theorem shows no prime power p^k with k≥2 can satisfy the first jump condition alone.","pith_inferences":["The same discriminant trick likely adapts to other even offsets: replacing 12 by 2m would produce a finite list of possible prime gaps from the factorizations of 8m, giving a path to similar characterizations for shifts like 6 or 20.","A natural testable prediction from the search is that any future solution beyond 10^12 will again be 65 (mod 72) and will factor as p(p+8); finding one that does not would refute the conjecture outright.","The conditional implication ties the infinitude of these solutions to the prime k-tuple conjecture for k=4; any unconditional proof of infinitely many such n would settle that case.","The elementary technique used to rule out prime powers might be extended to rule out numbers with exactly two prime factors of unequal size, though the semiprime case is already fully settled."],"forward_implications":["For squarefree semiprime n with n+12 semiprime, the simultaneous jump conditions hold only for the prime-quadruplet form n=p(p+8).","Every such semiprime solution is congruent to 65 modulo 72; under the open characterization conjecture, all composite solutions would share this residue.","No integer of the form p^k with k≥2 satisfies both jump conditions; in fact, the first condition alone fails for every prime power.","The exhaustive search up to 10^12 found exactly 166 composite solutions, all congruent to 65 (mod 72) and all of the prime-quadruplet semiprime type.","If the characterization conjecture is true and infinitely many solutions exist, then infinitely many prime quadruplets exist."],"fun_headline_variants":["Semiprime 12-jump solutions are exactly prime quadruplets","12-jump semiprime case reduces to prime quadruplet spacing","Squarefree semiprimes: 12-jump holds only for prime quadruplets","Prime quadruplets characterized among semiprime 12-jump solutions","Semiprime 12-jump iff prime quadruplet, no prime powers"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The completeness of the computational search up to 10^12 rests on an unverified custom segmented-sieve program; if that program has an overflow or sieving bug, the reported 166 solutions and the absence of counterexamples could be wrong.","fun_headline_variants_meta":{"raw":{"variants":["Semiprime 12-jump solutions are exactly prime quadruplets","12-jump semiprime case reduces to prime quadruplet spacing","Squarefree semiprimes: 12-jump holds only for prime quadruplets","Prime quadruplets characterized among semiprime 12-jump solutions","Semiprime 12-jump iff prime quadruplet, no prime powers"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000877,"raw_usage":{"total_tokens":3613,"prompt_tokens":710,"completion_tokens":2903,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":454,"completion_tokens_details":{"reasoning_tokens":2806}},"tokens_in":454,"tokens_out":2903,"duration_ms":23651,"temperature":1.0,"reasoning_tokens":2806,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-02T11:54:03.623599+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run an independent implementation of the same segmented sieve over n∈[2,10^12] and check that it returns exactly the same 166 solutions; any discrepancy would falsify the claimed completeness. Alternatively, a single composite n not of the form p(p+8) with (p,p+2,p+6,p+8) prime and satisfying both jump conditions would falsify the open conjecture, and a single prime power p^k satisfying φ(p^k+12)=φ(p^k)+12 would falsify Theorem 3.1.","supporting_citations":[],"review_version":2}