Researchers propose a self-distillation fix for ‘catastrophic forgetting’ in LLMs

LLMs tend to lose prior skills when fine-tuned for new tasks. A new self-distillation approach aims to reduce regression and ...

AI method accelerates liquid simulations by learning fundamental physical relationships

Researchers at the University of Bayreuth have developed a method using artificial intelligence that can significantly speed up the calculation of liquid properties. The AI approach predicts the ...

Sophelio Launches Data Fusion Labeler (dFL) to Accelerate Sensor Data Preparation for Machine Learning

Sophelio Introduces the Data Fusion Labeler (dFL) for Multimodal Time-Series Data - The only labeling and harmonization ...
chromatographyonline

A Novel Machine Learning Method for Predicting Retention Time of Small Molecule Pharmaceutical Compounds Across Reversed-phase HPLC Columns

Traditional QSRR models are limited to single-column predictions, hindering adaptability across diverse LC setups in pharmaceutical settings. The new ML-based approach predicts retention times using ...
EurekAlert!

Accelerating drug combination discovery with machine learning

(L) First and co-corresponding author Charlie Wright, PhD, St. and (R) co-corresponding author Paul Geeleher, PhD, both of the St. Jude Department of Computational Biology. (MEMPHIS, Tenn. – December ...
Frontiers

Machine learning methods for the detection and prediction of cognitive impairment in Parkinson’s disease: a systematic review and meta-analysis

Background: Cognitive impairment in Parkinson’s disease (PD-CI) is a prevalent non-motor symptom, significantly diminishing quality of life and imposing a substantial family burden. Effective ...
The Lancet

Effective sample size for individual risk predictions: quantifying uncertainty in machine learning models

Copyright: © 2025 The Author(s). Published by Elsevier Ltd. Individual prediction uncertainty is a key aspect of clinical prediction model performance; however ...
Frontiers

Comparison of multiple machine learning methods for predicting postoperative hyperglycemia in patients without diabetes undergoing cardiac surgery

Background: Stress-induced hyperglycemia (SHG) represents a significant metabolic complication in non-diabetic cardiac surgery older adult patients, with substantial implications for postoperative ...
National Academies of Sciences%2c Engineering%2c and Medicine

Making Machine Learning Safer in High-Stakes Settings

Machine learning, a key enabler of artificial intelligence, is increasingly used for applications like self-driving cars, medical devices, and advanced robots that work near humans — all contexts ...
PNAS

Machine learning in chemistry

The Perspective by Tiwary et al. (8) offers a comprehensive overview of generative AI methods in computational chemistry. Approaches that generate new outputs (e.g., inferring phase transitions) by ...
Hosted on MSN

How to Compare Machine Learning Models Using Bootstrap Resampling in Scikit-Learn

Learn how to compare ML models using bootstrap resampling with a hands-on sklearn implementation. Social Security, Medicare are "going to be gone," Donald Trump warns Here's What To Do If You See A ...