arXiv reader

On the largest prime factor of integers in short intervals III

Runbo Li

arXiv
2601.00910v1 · PDF · 1 January 2026
Categories
math.NT
Licence
CC BY SA 4.0 The site shows the text.

Abstract

Using Watt’s mean value theorem and a delicate sieve decomposition, the author shows that the interval [x, x+x^{\frac{1}{2}+\varepsilon}] contains an integer with a prime factor larger than x^{\frac{35}{36}-\varepsilon} for sufficiently large x. This gives a solution with \gamma = \frac{1}{36} to the Exercise 5.1 in Harman’s monograph and improves the previous record of the author proved in 2024, where \gamma = \frac{1}{26.5} is obtained.

Items 10

Theorem

Theorem 1

N4SV

For sufficiently large x, the interval [x, x+x^{\frac{1}{2}+\varepsilon}] contains an integer with a prime factor larger than x^{\frac{35}{36}-\varepsilon}.

Lemma

Lemma 2.1

JMRK

([[HeathBrownJia1718], Lemma 1]). Suppose that M N = v and v^{\frac{17}{35}} \ll M \ll v^{\frac{18}{35}}. Then for (\log x)^{2B} \leqslant T \leqslant x^{\frac{1}{2}-\frac{\varepsilon}{6}}, we have

\int_{T}^{2 T}|M(b+i t) N(b+i t) P^K (b+i t)| d t \ll (\log x)^{-B}.
Used by Proof
Lemma

Lemma 2.2

01ER

([[HeathBrownJia1718], Lemma 2]). Suppose that M N L=v, M \ll v^{\frac{18}{35}} and N \ll v^{\frac{9}{35}}. Then for \sqrt{L} \leqslant T \leqslant x^{\frac{1}{2}-\frac{\varepsilon}{6}}, we have

\int_{T}^{2 T}|M(b+i t) N(b+i t) L(b+i t) P^K (b+i t)| d t \ll (\log x)^{-B}.
Used by Proof
Lemma

Lemma 2.3

VMNZ

([[LRB122], Lemma 4.1]). Let T \geqslant 1, then

\int_{T}^{2 T}\left|L\left(\frac{1}{2}+i t\right)\right|^{4}\left|N\left(\frac{1}{2}+i t\right)\right|^{2} d t \ll \left(T+N^{2} T^{\frac{1}{2}}+ L^2 T^{-4} (N+T) \right) T^{\frac{\varepsilon}{2}}.
Used by Proof
Lemma

Lemma 2.4

4JEJ

Suppose that M N H L=v, and M, N and H satisfy one of the following 3 conditions:

(1). M \ll v^{\frac{18}{35}},\ N \gg H,\ N^{\frac{3}{4}}H \ll v^{\frac{9}{35}},\ N H^{\frac{1}{2}} \ll v^{\frac{9}{35}},\ N^{\frac{7}{4}} H^{\frac{3}{2}} \ll v^{\frac{18}{35}};

(2). M \ll v^{\frac{18}{35}},\ N \ll v^{\frac{9}{35}},\ H \ll v^{\frac{9}{140}};

(3). M \ll v^{\frac{18}{35}},\ N^2 H \ll v^{\frac{18}{35}},\ H \ll v^{\frac{9}{70}}.

Then for \sqrt{L} \leqslant T \leqslant x^{\frac{1}{2}-\frac{\varepsilon}{6}}, we have

\int_{T}^{2 T}|M(b+i t) N(b+i t) H(b+i t) L(b+i t) P^K (b+i t)| d t \ll (\log x)^{-B}.
Used by Proof
Proof of Lemma 2.4

We only need to prove that

\int_{T}^{2 T}\left|M\left(\frac{1}{2}+i t\right) N\left(\frac{1}{2}+i t\right) H\left(\frac{1}{2}+i t\right) L\left(\frac{1}{2}+i t\right) P^K\left(\frac{1}{2}+i t\right)\right| d t \ll x^{\frac{1}{2}} (\log x)^{-B}

holds true under each condition above.

Assume that Condition (1) holds. The proof of this case is similar to that of [[JiaLiu], Lemma 3] where [[DI84], Theorem 2] is used.

Assume that Condition (2) holds. Using Hölder’s inequality and the mean value theorem for Dirichlet polynomials, we have

\begin{aligned} & \int_{T}^{2 T}\left|M\left(\frac{1}{2}+i t\right) N\left(\frac{1}{2}+i t\right) H\left(\frac{1}{2}+i t\right) L\left(\frac{1}{2}+i t\right) P^K\left(\frac{1}{2}+i t\right)\right| d t \\ \ll&\ \left( \int_{T}^{2 T}\left|M\left(\frac{1}{2}+i t\right) \right|^2 \left| P^K\left(\frac{1}{2}+i t\right)\right|^2 d t \right)^{\frac{1}{2}} \left( \int_{T}^{2 T}\left|N\left(\frac{1}{2}+i t\right) \right|^4 d t \right)^{\frac{1}{4}} \\ &\ \left( \int_{T}^{2 T}\left|L\left(\frac{1}{2}+i t\right) \right|^4 \left|H\left(\frac{1}{2}+i t\right) \right|^4 d t \right)^{\frac{1}{4}} \\ \ll&\ x^{\frac{1}{4}} \left( \int_{T}^{2 T}\left|N\left(\frac{1}{2}+i t\right) \right|^4 d t \right)^{\frac{1}{4}} \left( \int_{T}^{2 T}\left|L\left(\frac{1}{2}+i t\right) \right|^4 \left|H\left(\frac{1}{2}+i t\right) \right|^4 d t \right)^{\frac{1}{4}}. \end{aligned}

By the mean value theorem for Dirichlet polynomials, we have

\left( \int_{T}^{2 T}\left|N\left(\frac{1}{2}+i t\right) \right|^4 d t \right)^{\frac{1}{4}} \ll \left(N^2 + T\right)^{\frac{1}{4}}.

Note that \sqrt{L} \leqslant T, we have

L^2 T^{-4} (N + T) \ll N + T.

Now, using Lemma 2.3 with N = H^2, we get

\left( \int_{T}^{2 T}\left|L\left(\frac{1}{2}+i t\right) \right|^4 \left|H\left(\frac{1}{2}+i t\right) \right|^4 d t \right)^{\frac{1}{4}} \ll \left(T + H^4 T^{\frac{1}{2}} + H^2 \right)^{\frac{1}{4}} T^{\frac{\varepsilon}{8}}.

Combining the bounds above with Condition (2), we have

\begin{aligned} & \int_{T}^{2 T}\left|M\left(\frac{1}{2}+i t\right) N\left(\frac{1}{2}+i t\right) H\left(\frac{1}{2}+i t\right) L\left(\frac{1}{2}+i t\right) P^K\left(\frac{1}{2}+i t\right)\right| d t \\ \ll&\ x^{\frac{1}{4}} \left(N^2 + T\right)^{\frac{1}{4}} \left(T + H^4 T^{\frac{1}{2}} + H^2 \right)^{\frac{1}{4}} T^{\frac{\varepsilon}{8}} \\ \ll&\ x^{\frac{1}{2}} (\log x)^{-B}. \end{aligned}

Note that this case is an improvement over Lemma 2.2.

Assume that Condition (3) holds. Using Hölder’s inequality and the mean value theorem for Dirichlet polynomials, we have

\begin{aligned} & \int_{T}^{2 T}\left|M\left(\frac{1}{2}+i t\right) N\left(\frac{1}{2}+i t\right) H\left(\frac{1}{2}+i t\right) L\left(\frac{1}{2}+i t\right) P^K\left(\frac{1}{2}+i t\right)\right| d t \\ \ll&\ \left( \int_{T}^{2 T}\left|M\left(\frac{1}{2}+i t\right) \right|^2 \left| P^K\left(\frac{1}{2}+i t\right)\right|^2 d t \right)^{\frac{1}{2}} \left( \int_{T}^{2 T}\left|N\left(\frac{1}{2}+i t\right) \right|^4 \left|H\left(\frac{1}{2}+i t\right) \right|^2 d t \right)^{\frac{1}{4}} \\ &\ \left( \int_{T}^{2 T}\left|L\left(\frac{1}{2}+i t\right) \right|^4 \left|H\left(\frac{1}{2}+i t\right) \right|^2 d t \right)^{\frac{1}{4}} \\ \ll&\ x^{\frac{1}{4}} \left( \int_{T}^{2 T}\left|N\left(\frac{1}{2}+i t\right) \right|^4 \left|H\left(\frac{1}{2}+i t\right) \right|^2 d t \right)^{\frac{1}{4}} \left( \int_{T}^{2 T}\left|L\left(\frac{1}{2}+i t\right) \right|^4 \left|H\left(\frac{1}{2}+i t\right) \right|^2 d t \right)^{\frac{1}{4}}. \end{aligned}

By Condition (3), Lemma 2.3 with N = H and the mean value theorem for Dirichlet polynomials, we have

\begin{aligned} & \int_{T}^{2 T}\left|M\left(\frac{1}{2}+i t\right) N\left(\frac{1}{2}+i t\right) H\left(\frac{1}{2}+i t\right) L\left(\frac{1}{2}+i t\right) P^K\left(\frac{1}{2}+i t\right)\right| d t \\ \ll&\ x^{\frac{1}{4}} \left(N^2 H + T\right)^{\frac{1}{4}} \left(T + H^2 T^{\frac{1}{2}} + H \right)^{\frac{1}{4}} T^{\frac{\varepsilon}{8}} \\ \ll&\ x^{\frac{1}{2}} (\log x)^{-B}. \end{aligned}

Now the proof of Lemma 2.4 is completed.

Lemma

Lemma 3.1

HUVI

We can give an asymptotic formula of the form (1) for

\sum_{t_1 \cdots t_n} S\left(\mathcal{A}_{p_1 \cdots p_n}, v^{\frac{1}{35}} \right)

and similar sums involving role-reversals if (t_1, \ldots, t_n) \in \mathcal{J}_{n}.

Proof of Lemma 3.1

This lemma can be proved by Lemma 2.4 together with the discussions in [[HeathBrownJia1718], Lemmas 7,9].

Lemma

Lemma 3.2

UP0G

We can give an asymptotic formula of the form (1) for

\sum_{t_1 \cdots t_n} S\left(\mathcal{A}_{p_1 \cdots p_n}, p_n \right)

and similar sums involving role-reversals if (t_1, \ldots, t_n) \in \mathcal{I}_{n}.

Proof of Lemma 3.2

This lemma can be proved by Lemma 2.1 together with the discussions in [[HeathBrownJia1718], Lemma 8].

Source and licence

The text on this page is from arXiv:2601.00910v1 by Runbo Li, under the licence CC BY SA 4.0. The corpus made these changes: it read the items out of the LaTeX source, it expanded the macros of the paper, it put each proof after its statement, and it converted the text to HTML. The formulas are the TeX of the paper, and KaTeX renders them. A formula or a reference can look different from the PDF. The page is published under CC BY SA 4.0.

The record of each item is in analysis/objects/2601/2601.00910.jsonl.