Back to search
2201.12750

Periodic Points and Arithmetic Degrees of Certain Birational Self-Maps

Long Wang

correctmedium confidence
Category
Not specified
Journal tier
Specialist/Solid
Processed
Sep 28, 2025, 12:56 AM

Audit review

The paper proves the bounded-height result by constructing, on a common resolution, a big divisor D = φ*H + ψ*H − 3π*H and deriving a two-sided height inequality h(f^n(R)) + h(f^{-n}(R)) ≥ A h(R) − C on a Zariski open set; Lemma 3.1 then yields bounded height for periodic points avoiding a proper closed Z. The construction uses Siu’s numerical criterion and Dang’s inequality to ensure bigness and thereby a uniform lower bound for the height associated to D off its base locus. This argument is complete and correct. By contrast, the model proposes a stronger one-sided inequality h_H(f^m(Q)) ≥ a_m h_H(Q) − C by asserting an unsupported decomposition ψ*H ≡ a_m π*H + E_m + N_m with E_m effective, which is generally false (pseudo-effectivity does not ensure an exact effective representative of a numerical class). Without establishing bigness and an appropriate lower bound off an augmented base locus, the claimed inequality and the resulting conclusion do not follow.

Referee report (LaTeX)

\textbf{Recommendation:} minor revisions

\textbf{Journal Tier:} specialist/solid

\textbf{Justification:}

The paper gives a clean, correct proof that periodic points of cohomologically hyperbolic birational self-maps have bounded height off a proper closed set, using a robust two-sided height inequality derived from a big divisor on a resolution. The argument is technically solid and well-situated in the literature. A few clarifications (explicitly stating the technical Lemma 3.1, the precise construction of Z, and a brief reminder of base-locus notions) would enhance readability, but the main results stand on firm ground.