A new security system called ClawSentry has been developed to protect large language model (LLM) agents from malicious attacks. The system uses a mult...
The Atom Learning Model (ALM) tokenises a school curriculum by breaking down math textbooks into individual 'atoms' - specific actions or steps that l...
Researchers have developed a new type of attack on visual world-model agents, such as DreamerV3. These attacks, called Critic-Induced Value-Subspace A...
Researchers have developed a modular agent that can reliably verify spatial relations in CT scans. The system works by breaking down the task into thr...
Researchers have proposed a new approach to developing artificial intelligence called Graph Engineering. This method involves creating dynamic graph s...
Researchers have proposed HIERA, a new approach to optimizing GPU kernel performance. Unlike existing methods that focus on a single implementation sp...
Researchers have developed AID-Guard, a protocol for stateful authorization in AI agents. The protocol ensures that only approved requests are execute...
Researchers have found that visual presentation can significantly impact the performance and failure modes of vision-language models (VLMs) in spatial...
Researchers have developed a new method for safe and efficient navigation in crowded areas with heterogeneous shapes. The Shape-Aware Reinforcement Le...
Researchers have developed a new method called DAMOS to assess speech quality. DAMOS uses machine learning to identify where distortions occur in spee...
Researchers have developed a new optimization model called Anchored Regularized Direct Least Squares (ARDLS) that addresses a key issue with the Analy...
Researchers have created a dataset of optical coherence tomography (OCT) images from guinea pigs with cochlear implants to study fibrosis and its impa...