Accurate water level forecasting is crucial for effective water resources management. Recently, a group of researchers from Jeonbuk National University has proposed a novel clustering-based machine learning framework for accurate, scalable, and computationally efficient water level prediction, especially in data-scarce regions. These findings are expected to lead to improved flood mitigation, agricultural irrigation, environmental sustainability, and long-term climate adaptation policies.

A new clinical trial led by the Alliance for Clinical Trials in Oncology will investigate if a combination of drug therapies after radiation therapy improves outcomes for people with newly-diagnosed, grade 3 IDH-mutant astrocytoma, a type of brain cancer. Supported in part by a grant from the National Cancer Institute, the study (Alliance A072301) will look at whether adding the oral medication vorasidenib to the standard oral chemotherapy can help keep the cancer from coming back after radiation.

A new study shows how fireflies speed up or slow down their flashing to sync up with other insects, creating a beautiful and other-wordly light show.

A new AI-driven system generates plans for long-term, complex tasks about twice as well as some existing methods. The technique, developed at MIT, uses two vision-language models that work together to simulate actions and produce files for formal planning software.

Researchers develop an axiomatic framework to clarify which risk-sharing rules follow from commonly desired principles in risk pools, such as anonymity of participant and incident information, non-punitive processes, and fairness. They characterize simple rules—including uniform, mean-proportional, and covariance-based linear rules—by formal properties like reshuffling, source-anonymous contributions, and aggregate contributions. The framework also defines broad rule classes and introduces scenario-based rules for settings where probabilistic modeling is impractical.

A new study presents a general, nonparametric statistical inference theory for integrals of quantiles without assuming a particular sampling design or dependence structure. The authors discuss conditions for consistency, bias, and asymptotic normality, and illustrate how the framework applies to widely used quantities in risk and inequality measurement, including tail-value-at-risk as well as Lorenz and Gini curves. Notes for dependent (time-series) data are also provided.