Researchers made AI agents better by turning academic papers into working code — no retraining required

ScholarEvolve uses published research to improve AI agents, raising Qwen's AppWorld task completion from 69% to 81.4% without retraining the model.

OpenAI Enhances Code Generation and Migration Performance for Low-Resource Languages like COBOL, and More | AI News Summary for October 20, 2026

AI news.OpenAI has announced that its latest models have significantly improved performance in programming languages that ...
Anthropic

Using Claude to reduce the attack surface of code

Automated attack surface analysis and reduction can meaningfully improve the security of deployed software in the near term.
BIOENGINEER

Extra-P Turns Scalability Modeling Into a Few Small Experiments

A decade-long effort at ETH Zurich and TU Darmstadt has distilled empirical performance modeling into a Python tool that ...
3D Printing Industry

Vibecoded clones of SolidWorks and AutoCAD released as open-source

Spend less time fixing files. Discover smarter workflows at AMA: Software 2026. Two open-source CAD applications written ...
BIOENGINEER

Rwanda’s cervical cancer screening gap defies near-universal awareness

A nationwide analysis of Rwanda's 2025 Demographic and Health Survey reveals that while 93.4% of women have heard of cervical ...
Business Outreach

AI Skills in 2026: 15 Skills You Need to Build a Career in Artificial Intelligence

Discover 15 essential AI skills in 2026, from Python and machine learning to generative AI and automation, to build a ...
Nature

Linear Models and Regression articles from across Nature Portfolio

Distributional regression (DR) refers to regression methods that model the entire conditional probability distribution of a response variable given a set of explanatory variables. The generalized ...
Science

Intron retention regulates STAT2 function and associates with immunotherapy response in lung cancer

Immunotherapy with checkpoint inhibitors has revolutionized cancer therapy and enabled considerable therapeutic advances for ...

The Trade-off with Explainability: The Difference Between a '95% Accurate Model That Never Gets Adopted' and a '60% Accurate System Used Every Day'

A machine learning model has 95% accuracy, yet it sits idle month after month. Meanwhile, a simple scoring system using a ...
HackerNoon

Why More AI Workloads Are Moving From the Cloud to Your Devices

A 10-step guide to on-device AI, covering model selection, quantization, deployment runtimes, hardware acceleration, hybrid ...