Scientists and engineers use simulators to model empirically observed phenomena. However, tuning the parameters of a simulator to ensure its outputs match observed data presents a significant challenge. Simulation-based inference (SBI) addresses this by enabling Bayesian inference for simulators, identifying parameters that match observed data and align with prior knowledge. Unlike traditional Bayesian inference, SBI only needs access to simulations from the model and does not require evaluations of the likelihood-function. In addition, SBI algorithms do not require gradients through the simulator, allow for massive parallelization of simulations, and can perform inference for different observations without further simulations or training, thereby amortizing inference. Over the past years, we have developed, maintained, and extended $\texttt{sbi}$, a PyTorch-based package that implements Bayesian SBI algorithms based on neural networks. The $\texttt{sbi}$ toolkit implements a wide range of inference methods, neural network architectures, sampling methods, and diagnostic tools. In addition, it provides well-tested default settings but also offers flexibility to fully customize every step of the simulation-based inference workflow. Taken together, the $\texttt{sbi}$ toolkit enables scientists and engineers to apply state-of-the-art SBI methods to black-box simulators, opening up new possibilities for aligning simulations with empirically observed data.

Large Language Models (LLMs) have gained significant popularity in recent years for their ability to answer questions in various fields. However, these models have a tendency to "hallucinate" their responses, making it challenging to evaluate their performance. A major challenge is determining how to assess the certainty of a model's predictions and how it correlates with accuracy. In this work, we introduce an analysis for evaluating the performance of popular open-source LLMs, as well as gpt-3.5 Turbo, on multiple choice physics questionnaires. We focus on the relationship between answer accuracy and variability in topics related to physics. Our findings suggest that most models provide accurate replies in cases where they are certain, but this is by far not a general behavior. The relationship between accuracy and uncertainty exposes a broad horizontal bell-shaped distribution. We report how the asymmetry between accuracy and uncertainty intensifies as the questions demand more logical reasoning of the LLM agent, while the same relationship remains sharp for knowledge retrieval tasks.

Electron beam accelerators are essential in many scientific and technological fields. Their operation relies heavily on the stability and precision of the electron beam. Traditional diagnostic techniques encounter difficulties in addressing the complex and dynamic nature of electron beams. Particularly in the context of free-electron lasers (FELs), it is fundamentally impossible to measure the lasing-on and lasingoff electron power profiles for a single electron bunch. This is a crucial hurdle in the exact reconstruction of the photon pulse profile. To overcome this hurdle, we developed a machine learning model that predicts the temporal power profile of the electron bunch in the lasing-off regime using machine parameters that can be obtained when lasing is on. The model was statistically validated and showed superior predictions compared to the state-of-the-art batch calibrations. The work we present here is a critical element for a virtual pulse reconstruction diagnostic (VPRD) tool designed to reconstruct the power profile of individual photon pulses without requiring repeated measurements in the lasing-off regime. This promises to significantly enhance the diagnostic capabilities in FELs at large.