This session brings together some of the most recent measures of evidence introduced in the Bayesian statistical literature.
These measures, specifically designed to solve problems of sharp or precise statistical hypotheses, are based on the idea of quantifying the discrepancy between the a posteriori distribution and the hypothesis of interest. This quantification can be done in various ways, e.g. by using the posterior median as a benchmark in univariate cases or by examining the 'tangent set', i.e. the set with the highest density under the null hypothesis. This session aims to show the adaptability of these measures by highlighting recent advances in both theoretical developments and real-world applications. Contributions range from the use of these measures in scenarios involving multi-parametric and non-standard models, showing their wide utility and relevance.