An arXiv preprint reports an if-and-only-if test for compactness in a stated class of radial weighted Bergman spaces. For a single Toeplitz operator with a bounded symbol, compactness holds exactly when the operator's Berezin transform vanishes at the boundary. The result is conditional on the paper's admissible radial weight class.
The question is whether boundary behavior of the Berezin transform can fully characterize compactness for one operator. The study analyzes mathematical inputs on the unit disk and does not define an empirical participant or observational sample.
The setting is a radial weighted Bergman space, a function space built from mathematical inputs on the unit disk, with the weight restricted to the radial class used in the paper. The operators are formed from bounded symbols. That scope matters: the theorem is about a defined mathematical family, not arbitrary elements of the Toeplitz algebra.
The proof turns functions into a matrix
To prove the result, the authors use a polynomial frame, meaning a collection of polynomial building blocks for representing the space, and analyze the infinite matrix representation it gives the Toeplitz operator. The paper also states that its family of dyadic polynomial functions is a Parseval frame for the weighted Bergman space.
The matrix analysis is divided into diagonal, lower-triangular and upper-triangular parts. The proof then localizes these pieces. An appendix proves a Carleson embedding estimate used by the main argument.
The boundary-vanishing condition allows the matrix operator to be approximated in operator norm by finite-rank operators. In other words, the infinite matrix can be replaced, as closely as required in operator norm, by an operator of finite rank. This is the proof statement that supports compactness.
For the necessity direction, the authors use normalized reproducing kernels, test functions associated with points in the space. They state that these kernels converge weakly to zero as their points approach the boundary. This links compactness to the vanishing of the Berezin transform near the disk's edge.
The abstract says the argument is new even in the unweighted Bergman space and does not rely on classical translation operators. The paper presents this as a translation-free route to the single-operator characterization.
Where the test stops
The paper then gives a warning about products. It constructs a product of two Toeplitz operators that is noncompact even though its Berezin transform vanishes at the boundary. A criterion that is exact for one operator therefore cannot simply be assumed for a product.
The main theorem is restricted to a single Toeplitz operator in the Hilbert-space setting, and the authors say the method does not directly extend to arbitrary elements of the Toeplitz algebra. The product example is a constructed counterexample, not a finding that every product behaves in the same way.
The supplied document is an arXiv preprint labeled version 2 and dated 29 August 2026. The authors report support from the National Natural Science Foundation of China, grant No. 12471116, and 2025CDJ-IAIS YB-004 at Chongqing University.
Its AI Use Disclosure says the authors developed the mathematical ideas and verified the arguments, while AI tools helped with drafting, language editing and consistency checks. The authors say they take full responsibility.
Paper data and sources
Original title: Compact Toeplitz operators on radial weighted Bergman spaces
Authors: Yuerang Li, Zipeng Wang
Journal/Repository: arXiv
Status: Preprint, not yet peer-reviewed
First online: 2026-08-24
DOI: Not available
Original paper · Full text