ICM 2026 Invited Section Lecture - Hong Wang

Value of the Information & Strength of the Argument

The lecture provides a clear, high-level survey of a frontier research area, bridging harmonic analysis and geometric measure theory. The argumentation is logically structured: it starts from classical Fourier uncertainty, builds wave packet decompositions, and reduces both restriction and projection problems to discrete incidence estimates. The high-low method is justified through concrete frequency localization and a simple geometric observation about the overlap of dual boxes. The explanations are conceptually strong, though the transcript lacks some technical details and the exact statements of the theorems, which limits strictly self-contained verification.

Scientific Rigor, Source Quality, Title Accuracy

The talk demonstrates high scientific rigor, with explicit references to classical theorems (e.g., Riesz, Fefferman, Bourgain, Szemerédi–Trotter) and to recent work by Roth, Guth, Solomon, and Wang. The approach is consistent and the statements are made with appropriate caveats, such as the distinction between sharp and smoothed cutoffs. The title, however, is excessively general and does not hint at the specific focus on projection theorems, which slightly reduces its informative value.

The title is a generic lecture label; it accurately identifies the speaker and event but does not convey the specific topic of projection theorems.

9/10

Presentation by an invited ICM speaker, with precise mathematical statements and references to classical and recent results. The reasoning is rigorous and self-contained, although the visual material is not available in the transcript and some references are only named informally.

Markers derived by PSI from the transcript: the creator did not define chapters.

• 00:00Introduction and outline: relation between restriction theory, projection estimates, and incidence for tubes.
• 03:00Fourier transform basics, uncertainty principle, and construction of smooth bump functions on convex sets.
• 07:00Dual convex sets, rectangles, and explicit calculation of the Fourier transform of indicator-like bump functions.
• 11:00Restriction problem: statement, threshold exponent, and failure of convergence on the sphere; role of concave manifolds.
• 15:00Wave packet decomposition via curved pieces and dual tubes; illustration with sphere and cone.
• 20:00Incidence for tubes: definition, relation to Kakeya sets, and distinction from point-line incidence geometry.
• 25:00Connection to orthogonal projections: covering numbers, Hausdorff dimension, and translation to incidence problems.
• 30:00High-low method: frequency decomposition, geometric overlap argument, and resulting incidence bound.
• 35:00Discussion of generalizations, connections with nonlinear projections, and open problems.

• Riesz theorem (1920s) - mentioned for dimension-one convergence — Referenced as the classical result for Fourier series convergence in one dimension.
• Fefferman - disproof of the ball multiplier conjecture — Mentioned to explain the failure of convergence for high dimensions.
• Szemerédi–Trotter theorem — Cited as a classical incidence bound for points and lines in the plane.
• Guth, Solomon, Wang (2019) - high-low method — Presented as the reintroduction of Roth's high-low method for incidence problems.