Answers are grounded in this paper and cite their sources.
We check Sections 2.4, 3 and 4 (the permutation and F_q[t] analogues) of A. Granville, D. Koukoulopoulos and J. Maynard, "Sieve weights and their smoothings" (Ann. Sci. Éc. Norm. Supér. (4) 54 (2021), 1089–1177). Numbering follows the final arXiv version, arXiv:1606.06781v4. (i) The divisor sum M(c;m) in (3.1) omits the multiplicity ∏ binom(c_j, b_j); as printed, Proposition 3.1 fails (for m=1 the right-hand side is 1−1/e for every k, whereas c(1,k) is the Bell number B_2k). We restore the proposition with a short generating-function proof, which also gives a Pascal-type shift identity and a two-line replacement for the end of the proof of Theorem 1.2. (ii) The expansion binom(q,c_1)/(q−1)^(c_1) = (1/c_1!)(1+O(c_1/q)) claimed for c_1 ≤ q is false beyond c_1 ≈ √q; a quadratic Poisson moment repairs the error term, and Theorem 1.2 holds as stated. (iii) In the saddle-point heuristic of Section 2.4 for Poly_q(n,m,2;k), three slips give the exponent 2^(2k−1)−6k−1; corrected, it is 2^(2k−1)−4k−1. We prove that the q→∞ limit is a 2k-fold mixed difference of a lattice-point count, obtain 15m²−51m+65 for k=2, and estimate the exponent for k=3 numerically at 20–22 (an extrapolation, not a proof). The qualitative conclusions of the paper are unaffected. The record includes the note (PDF and LaTeX), exact-arithmetic verification code, data and the figure.
No takes yet. Share an insight, caveat, or question.
Ruqing Chen (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: