Researchers show how explainable AI can pinpoint and break LLM safety filters
Researchers used explainable AI to fingerprint the internal layers responsible for safety alignment in open-source large language models and showed ...
Researchers used explainable AI to fingerprint the internal layers responsible for safety alignment in open-source large language models and showed ...
A new narrative review in Discover Mental Health documents nineteen publicly reported cases in which sustained emotional reliance on AI ...
Researchers at King Abdulaziz University used a multi-model large language framework with human oversight to develop and preliminarily validate a ...
A schema-enforced large language model framework achieved 94 percent reproducibility across simulated sarcoma tumor board decisions, a leap over earlier ...
A new PLOS Climate experiment finds that ChatGPT's credibility on climate claims depends more on users' prior beliefs than on ...
Greek researchers have built a fully open-source retrieval-augmented chatbot that answers journalists' questions using 73,000 archived Greek news articles, complete ...
A new bibliometric analysis of 779 publications shows AI in science education research exploding since 2023, while biology and geology ...
Researchers at the University of Naples Federico II have built a hybrid AI system that combines ontology-driven knowledge graphs with ...
Researchers developed EntQA, a reference-free, entity-centric metric that scores how faithfully large language models preserve clinically relevant concepts in medical ...
A large experiment around the 2024 UK election found that researching political claims with AI chatbots increased accurate knowledge as ...
© 2026 Scienmag - Science Magazine
© 2026 Scienmag - Science Magazine