This paper introduces a novel simulation-based inference framework to constrain cosmological and astrophysical parameters from the Lyman-alpha forest 1D power spectrum using CAMELS hydrodynamic simulations. It demonstrates excellent performance when trained and tested on the same galaxy formation model, and proposes multi-domain training as an effective solution to overcome biases when dealing with different models.

This paper investigates how to effectively utilize Vision-Language Models (VLMs) for image forgery detection and localization (IFDL). It introduces IFDL-VLM, a novel two-stage pipeline that decouples the core IFDL task from VLM-based explanation generation, leveraging localization masks to significantly enhance VLM interpretability and achieving state-of-the-art results across multiple benchmarks.

This paper introduces a teaching case that uses generative AI personas in a stakeholder simulation to educate students on Agile Requirements Engineering. It aims to provide practical experience in requirements elicitation and documentation while fostering critical reflection on the limitations and ethical considerations of AI.

This paper introduces a novel PCA sweep procedure for Supervised Semantic Differential (SSD), a method modeling how text meaning varies with individual differences. The sweep systematically selects the optimal number of PCA components to ensure interpretable and stable semantic gradients, illustrated through a case study on AI discourse related to narcissism.

This paper introduces the Memory Printer, a tangible device combining slow design and generative AI for reconstructing unrecorded personal memories. It explores how embodied, layered interactions can enhance user agency and emotional engagement in reminiscing, while also highlighting critical concerns regarding false memories, algorithmic bias, and data privacy in emotionally sensitive AI applications.