The authors study recent improvements of black-box VI and study the effect of step-size search, the sticking the landing (STL) gradient estimator, using importance weighting in the training objective and/or at test time, and the use of RealNVP flows as a more flexible approximating posterior. Combining these significantly outperforms naive ADVI. Strengths: - the empirical evaluation is careful and convincing - 30 benchmarks were considered Weaknesses: - no new method was proposed