The uniqueness theorem for Kasparov theory
Gábor Szabó · 2026
Read The uniqueness theorem for Kasparov theory by Gábor Szabó (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-23029.txt.
Introduction
Gábor Szabó proves a general uniqueness theorem for Kasparov theory, a tool in operator algebra. The result answers a question raised by Carrión and co-authors in a landmark paper on classifying C-star algebras. Uniqueness theorems say when two constructions must give the same answer.
Given two separable algebras and a pair of absorbing representations, the induced KK-theory element vanishes under the stated conditions. Such vanishing results are the engine behind classification programs for algebras. They tell mathematicians when deep structural questions have a single possible answer.
Catalog record
| CREATOR | Gábor Szabó |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 109 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2601-23029
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
The uniqueness theorem for Kasparov theory Thanks: Funded by the European Union. Views and opinions expressed are those of the authors only and do not necessarily reflect those of the European Union or the European Research Council. Neither the EU nor the ERC can be held responsible for them. Gábor Szabó Abstract . Answering a question of Carrión et al in their recent landmark paper on C ∗ \mathrm{C}^{*} -algebra classification, we prove a general uniqueness theorem for K K KK -theory. Given arbitrary separable C ∗ \mathrm{C}^{*} -algebras A A and B B and a Cuntz pair consisting of two absorbing representations φ , ψ : A → ℳ ( B ⊗ 𝒦 ) \varphi,\psi:A\to\mathcal{M}(B\otimes\mathcal{K}) , the induced element of K K ( A , B ) KK(A,B) vanishes if and only if φ \varphi and ψ \psi are strongly asymptotically unitarily equivalent. This improves upon the Lin–Dadarlat–Eilers stable uniqueness theorem. The conclusion is deduced by first showing the K 1 K_{1} -injectivity of an auxiliary C ∗ \mathrm{C}^{*} -algebra associated to the C ∗ \mathrm{C}^{*} -pair ( A , B ) (A,B) , which is sometimes called the Paschke dual algebra in the literature. Most of the article is concerned with the treatment of an umbrella theorem, which yields such a uniqueness theorem for other variants of K K KK -theory. This encompasses nuclear K K KK -theory, ideal-related K K KK -theory, equivariant K K KK -theory, or any combinations thereof. 2020 Mathematics Subject Classification 19K35, 46L55, 46L35 Introduction Kasparov theory, also known as KK -theory, is a very powerful invariant for \mathrm{C}^{*} -algebras. It is a bivariant version of topological K -theory, which includes ordinary K -theory and K -homology, alongside with Brown–Douglas–Fillmore theory [ 5 ] , as special cases. Invented by Kasparov [ 30 , 31 ] , it can be viewed as either a bifunctor from (separable) C*-algebras to abelian groups, or as a way to equip the class of (separable) \mathrm{C}^{*} -algebras with a more flexible notion of homomorphisms. In fact, given a second-countable locally compact group G , Kasparov’s construction also yields such a bifunctor on the category of all (separable) G - \mathrm{C}^{*} -algebras, which provided impactful applications on classical problems such as the Novikov conjecture. The original approach by Kasparov, commonly referred to as the Kasparov picture, is still the most common way of treating this subject, but there are other equivalent appproaches that are geared towards different types of applications. In this article we are mainly interested in the Cuntz picture [ 8 ] and the Cuntz–Thomsen picture [ 42 ] for (equivariant) KK -groups. In Cuntz’ framework, elements in Kasparov’s group KK(A,B) for separable \mathrm{C}^{*} -algebras A and B can be described as homotopy classes of so-called Cuntz pairs , i.e., pairs of * -homomorphisms \varphi,\psi:A\to\mathcal{M}(B\otimes\mathcal{K}) such that \varphi(a)-\psi(a)\in B\otimes\mathcal{K} for all a\in A . For equivariant KK -theory, one considers pairs of cocycle representations satisfying a similar condition; see also [ 41 , Section 6] . A groundbreaking theorem concerning the analytical properties of KK -theory is the so-called stable uniqueness theorem proved independently by Lin [ 34 ] and Dadarlat–Eilers [ 12 , 13 ] . It states that if a Cuntz pair (\varphi,\psi) represents the zero element in KK(A,B) , then \varphi and \psi are homotopic via conjugating one of the maps with a unitary path in \mathbf{1}+B\otimes\mathcal{K} after stabilizing both \varphi and \psi with a so-called absorbing representation. Note that an elegant treatment of the stable uniqueness theorem was recently included in [ 10 ] . Over the years, this theorem has become one of the central pieces of methodology in the structure and classification theory of nuclear \mathrm{C}^{*} -algebras; see for example [ 46 , 26 , 25 , 16 , 14 , 22 , 38 , 23 , 18 , 24 , 6 ] . More recently, I have generalized the stable uniqueness theorem to the setting of \mathrm{C}^{*} -dynamics together with Gabe in [ 21 ] , which was one of the key conceptual ingredients behind our dynamical Kirchberg–Phillips theorem [ 20 ] . Further interesting applications of this theorem may be expected to come in the realm of \mathrm{C}^{*} -dynamics and their classification. Its counterpart for actions of unitary tensor categories followed soon afterwards [ 33 ] and may become instrumental to classify unitary tensor category actions on Kirchberg algebras. When one considers the stable uniqueness theorem (either in its \mathrm{C}^{*} -algebraic or \mathrm{C}^{*} -dynamical form), the role of the absorbing representation is that of a necessary largeness criterion for the representations in question. The notion of an absorbing representation — recalled after subsection 1.1 for \mathrm{C}^{*} -dynamics — is at the same time inspired by the conclusion of Voiculescu’s fundamental theorem [ 47 ] . It can be interpreted as asserting that every unitizably full 1 1 1 In this specific context, this can be interpreted as one of two more familiar statements. If the \mathrm{C}^{*} -algebra is non-unital, then the representation needs to be faithful and essential. If the \mathrm{C}^{*} -algebra is unital, then it also needs to have the property that the representation is supported on a subspace with infinite-dimensional orthogonal complement. representation of a separable \mathrm{C}^{*} -algebra on the Hilbert space \ell^{2}(\mathbb{N}) is absorbing. It is known that all elements of a given KK -group can be represented by a Cuntz pair of absorbing representations. When one is faced with applications to the classification of nuclear \mathrm{C}^{*} -algebras, such an absorbing behavior can be detected via Elliott–Kucerovsky’s “purely largeness” criterion [ 15 ] and its known variants; see also [ 3 , Subsection 4.2] . For the purpose of this article, one of the primary applications to have in mind in classification is the approach of Carrión et al [ 6 ] , where part of the technical work is about getting closer to a setup where one can compare a pair of naturally occurring absorbing representations via KK -theory. The stable uniqueness theorem allows one to obtain a strong form of equivalence for the pair of representations, but in general only after first stabilizing with an auxiliary representation. The process to “destabilize”, so to speak, may require a substantial amount of extra work involving the specific assumptions related to the application. Schafhauser demonstrated in [ 38 ] that this issue can be circumvented under a UHF-stability assumption. Building on this idea, it is shown in [ 6 , Theorem 5.15] that one can omit the stabilization with the auxiliary representation at the expense of adding a tensor copy of the Jiang–Su algebra \mathcal{Z} to the codomain. Since said article concerns the classification of \mathrm{C}^{*} -algebras that are ultimately Jiang–Su stable, this extra tensor copy of \mathcal{Z} can be removed in a later phase of the application with some extra steps. However, it is a natural question to what extent it is really necessary to stabilize with an auxiliary representation when one compares a pair of representations that are absorbing to begin with. This was asked as [ 6 , Question 5.17] and [ 39 , Problem LXII] . One would refer to the desired statement below