About
Research & Development
Securing AI Agents: Addressing Time-of-Use Vulnerabilities in LLM Workflows
TLDR: A new research paper introduces the first study on Time-of-Check to Time-of-Use (TOCTOU) vulnerabilities in LLM-enabled agents. These vulnerabilities arise when an agent...
Research & Development
Orthogonal Natural Gradient Descent: A New Approach to Continual Learning
TLDR: Researchers introduced Orthogonal Natural Gradient Descent (ONG), combining natural gradients (which account for parameter space geometry) with orthogonal gradient descent (which prevents forgetting...
Research & Development
BudgetThinker: Guiding Language Models to Reason Efficiently within Set Limits
TLDR: BudgetThinker is a novel framework that enables Large Language Models (LLMs) to perform complex reasoning while strictly adhering to a specified token budget....
Research & Development
Enhancing Speech Restoration with Multi-Metric Preference Alignment
TLDR: A new research paper introduces a multi-metric preference alignment strategy for generative speech restoration. By creating a unique dataset (GenSR-Pref) based on unanimous...
Research & Development
Mimicking Human Vision: A Deep Network for Perceiving Hidden Shapes
TLDR: Researchers have developed ICPNet, a novel deep neural network inspired by the human visual cortex, to enable machines to perceive "illusory contours" –...
Research & Development
Unlocking Transformer Generalization: The Power of Attention’s Effective Rank
TLDR: A new research paper by Aviral Dhingra proposes that the generalization ability of overparameterized neural networks, especially Transformers, can be explained by the...