Academic papers, research breakthroughs, and scientific advances in machine learning and AI.
204 articles
Machine-learning models now enable near ab initio simulations of complex molecular and condensed-phase systems. At the same time, accurately describing... Source: publishing.aip.org
Researchers developed an AI-powered speech neuroprosthesis that decodes high-density intracortical signals into fluent, high-fidelity synthetic speech. Source: neurosciencenews.com
July 17, 2026 — For grid-scale energy storage and national energy resilience, the U.S. needs better batteries. Lawrence Livermore National Laboratory (LLNL)... Source: hpcwire.com
The history of drug discovery is littered with confident target calls that failed in the clinic, and artificial intelligence (AI) target identification has... Source: drugdiscoverynews.com
Can a large language model (LLM) improve at code generation using only its own raw outputs, without a verifier, a teacher model, or… Source: machinelearning.apple.com
A powerful new AI tool has uncovered what could be one of the biggest integrity problems in modern science. After analyzing 2.6 million cancer research... Source: sciencedaily.com
This research utilizes Quantum Generative Adversarial Networks, or QGANs, a promising architecture in the field, to assess post-quantum cryptographic... Source: quantumzeitgeist.com
Large Language Models (LLMs) are increasingly deployed to autonomously solve real-world tasks. A key ingredient for this is the LLM… Source: machinelearning.apple.com
For grid-scale energy storage and national energy resilience, the U.S. needs better batteries. Lawrence Livermore National Laboratory (LLNL) scientists are... Source: techxplore.com
Study finds new equation matches accuracy of the original and is available for laboratories to use at no cost through a transparent and single line of code. Source: hopkinsmedicine.org
Asthma caused around 436000 deaths globally in 2021. Predicting asthma attacks can save lives, reduce costs, and strengthen healthcare systems. Source: nature.com
A new study reveals a noninvasive BCI framework that aligns human neuroplasticity with AI to match invasive accuracies. Source: neurosciencenews.com
Researchers say the simplified Martin-Hopkins equation could help clinicians more accurately identify patients who need cholesterol-lowering therapies. Source: insideprecisionmedicine.com
Researchers develop BINND, a deep learning model that predicts complex DNA-DNA binding with 83.5% accuracy, unlocking scalable DNA computing. Source: neurosciencenews.com
Pilot randomized trial finds EHR-integrated decision support improved clinicians' prediction of school-age asthma. A machine learning tool that analyzes... Source: regenstrief.org
The new hybrid framework aims to balance detail, spatiotemporal coverage, and physical interpretability in city-scale weather and environment predictions. Source: anl.gov
Researchers analyzed biomarkers that could help psychiatrists identify at-risk individuals earlier and more objectively, supplementing existing self-report... Source: viterbischool.usc.edu
A groundbreaking study published in Nature Communications in 2026 sheds new light on the persistent challenge of bias in the emulation of target trials for... Source: bioengineer.org
Researchers from Rotonium, Centre for Quantum Technologies at National University of Singapore, Inveriant, Politecnico di Milano, and CNIT published a... Source: semiengineering.com
Loop engineering explained through Karpathy's autoresearch repo and Bilevel Autoresearch paper, covering verifiers, state, stop conditions. Source: marktechpost.com