On a Reinforcement Learning Methodology for Epidemic Control, with application to COVID-19

A new study proposes the mean square information criterion (MIC) for order selection in vector autoregressive (VAR) models, addressing limitations of existing methods like AIC, BIC, and Hannan-Quinn criteria. The research indicates that MIC can consistently estimate the true order of VAR processes under mild conditions, outperforming traditional methods, especially in smaller dimensions.

A recent study investigates the capability of large language models (LLMs) to detect misinformation in scientific news reporting, particularly in the context of the COVID-19 pandemic. The research introduces a new dataset, SciNews, comprising 2.4k scientific news stories from both trusted and untrusted sources, aiming to address the challenge of misinformation without relying on explicitly labeled claims.

A novel framework for measuring historical structural oppression has been introduced, utilizing Large Language Models (LLMs) to generate context-sensitive scores of lived historical disadvantage across various geopolitical settings. This approach addresses the limitations of traditional measurement methods that often overlook identity-based exclusion and rely heavily on material resources.

A new Bayesian model has been developed to address the challenges of multi-stage censoring in healthcare decision-making, particularly in oncology. This model aims to improve risk estimation by accounting for the selective censoring of outcomes, which often affects underserved patient groups. The model's effectiveness was demonstrated in synthetic settings, showcasing its potential to recover true outcomes despite data limitations.