Researchers have developed a framework to help vision-language models better connect words with the visual elements they describe, improving their ability to understand unseen combinations of objects and concepts.
New research details an AI-powered browser extension that dynamically tailors educational examples to individual student needs and real-time interaction.
Researchers have developed a system that enforces data isolation in AI planning workflows, enhancing transparency and auditability.
A new method allows large language models to process information and generate responses concurrently, dramatically reducing latency without sacrificing accuracy or safety.
Researchers have developed a framework that goes beyond simple keyword matching to identify and restore subtly altered phrases used in potential academic misconduct.
A new framework, UniCoR, significantly improves code retrieval by learning to understand code’s meaning across different programming languages.
New research shows that large language models are surprisingly adept at recognizing and applying complex grammatical rules without explicit programming.
New research introduces a rigorous benchmark and optimization techniques for improving the ability of advanced visual AI models to distinguish subtle differences in images.
Researchers have achieved nanoscale nuclear magnetic resonance using a silicon carbide-based spin sensor, opening doors to atomic-resolution imaging of molecular structures.
A new reinforcement learning algorithm, Symphony, dramatically improves the sample efficiency and safety of humanoid robot locomotion.