Fractional-polynomial prime-pair correlation conjecture

Let f(x)=a1xc1+a2xc2++akxckf(x)=a_1x^{c_1}+a_2x^{c_2}+\ldots+a_kx^{c_k}, where A>1A>1, c1>0c_1>0 is a non-integer, a10a_1\ne0, and c1>c2>>ckc_1>c_2>\ldots>c_k. For NN sufficiently large, set

M=N1/c1logBN.M=\frac{N^{1/c_1}}{\log^B N}.

Fractional-polynomial prime-pair conjecture. There exists B=B(f,A)B=B(f,A) such that

EmMEnNΛ(n)Λ(n+f(m))=1+O(1logAN).\mathop{\mathbb{E}}_{m\leqslant M}\mathop{\mathbb{E}}_{n\leqslant N}\Lambda(n)\Lambda\bigl(n+\lfloor f(m)\rfloor\bigr)=1+O\left(\frac{1}{\log^A N}\right).

This extends the paper's result to all fractional polynomials, including the previously open range 1<c<21<c<2. The proposed approach would require a Vinogradov mean value theorem for fractional polynomials; the authors also expect analogous results for suitable Hardy-field functions.

Sources & referencesView supporting material

Primary source

Bora Çalım, Ioannis Iakovakis, Sophie Long, Jack Moffatt and Deborah Wooton, “Popular differences in primes along fractional powers”, arXiv:2411.17599 (2024).

Additional references

2 papers in this index state this conjecture (2017–2024). The statement above is taken from the most recent of them; the others are arXiv:1708.04841.

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.