Friday, 8 May, 2026 
at Anton Rauch Lecture Hall, Atominstitut, Stadionallee 2, 1020 Wien

Host: Markus Müller

Programme:

10:30 Get Together with Snacks

11:00 VCQ Student Talk by Manuel Mekonnen

11:15 VCQ Colloquium Talk Ludovico Lami

You can follow this talk via zoom here (passcode vcq_2026).

Talk details:

On the ultimate limits of entanglement testing

The task of entanglement testing is to discriminate a given entangled state from the set of all separable (i.e. unentangled) states. Two types of error are possible: type I, where one mistakes an entangled state for a separable one, and type II, where one makes the opposite error. In the asymptotic regime, in which one is given many copies of the state to test, the probabilities of these errors will typically decay exponentially to zero. The ultimate limits of entanglement testing are therefore quantified by the optimal type I and type II error exponents governing this convergence. These can be determined with the help of two key results in entanglement theory: the generalised Sanov theorem and the generalised quantum Stein’s lemma, respectively. The latter has a long history, beginning with a claimed proof in [Brandão/Plenio, CMP 295:791, 2010], which was later found to contain a fatal gap [Berta et al., Quantum 7:1103, 2023; Berta et al., Nat. Phys. 20:172, 2024]. I will present the motivation behind these results, highlight their connections with the theory of entanglement manipulation and the notion of asymptotic reversibility, explain their essential content in a mostly nontechnical and operational way, and survey the recent proofs of these statements in [Lami/Berta/Regula, Nat. Phys. 22:439, 2026] and [Lami, IEEE Trans. Inf. Theory 71:4454, 2025].