Decoding Conspiracy: How AI Agents Spot the Markers of Belief
Researchers are harnessing the power of artificial intelligence to identify the subtle linguistic cues that indicate belief in conspiracy theories and understand how these ideas spread.
Science
Smart Surfaces, Smarter Bidding: Optimizing Wireless Networks with AI
A novel approach leverages reinforcement learning and auction theory to dynamically allocate reconfigurable intelligent surfaces for enhanced spectral efficiency and cost control.
Wind Turbine Forecasts Get a Boost from Collaborative Learning
![A system of four hundred turbines, sampled via nearest-neighbor analysis, demonstrates daily generation fluctuations, and subsequent forecasting leverages federated clustering combined with cluster-specific federated [latex]LSTM[/latex] models to anticipate these variations-a methodology acknowledging inherent systemic drift and emphasizing localized prediction within a distributed network.](https://arxiv.org/html/2603.05263v1/2603.05263v1/clustering/visualization.png)
A new framework leverages the power of federated learning and behavioural analysis to improve the accuracy and scalability of wind power forecasting for distributed energy systems.
Seeing is Believing? New Method Flags Vision AI’s Hallucinations Before They Happen
Researchers have developed a novel framework that predicts when vision-language models are likely to generate inaccurate or fabricated descriptions, offering a path toward more reliable AI systems.
Seeing Clearly: AI Restores Detail to Ground-Based Solar Images
A new physics-informed neural network approach overcomes atmospheric distortion to reveal finer details in ground-based observations of the Sun.
Seeing Through the Complexity: AI and Brain Glioma Detection
This review examines how artificial intelligence, particularly deep learning, is transforming the analysis of brain gliomas from MRI scans.
Echoes of Training: Unmasking Data Used to Build AI Models
![The method quantifies gradient deviation to assess the sensitivity of a model’s output to perturbations in its input, effectively measuring the extent to which small changes can induce significant alterations-a principle formalized as [latex] \delta y = \frac{\partial y}{\partial x} \delta x [/latex]-and thereby providing a robust indicator of model stability and reliability.](https://arxiv.org/html/2603.04828v1/2603.04828v1/x6.png)
Researchers have developed a new technique to identify whether specific text samples were used in the pre-training of large language models.
Unlocking Vision Transformer Insights: Probing for Out-of-Distribution Generalization
New research reveals that carefully examining the internal layers of Vision Transformers-specifically within their feedforward networks-offers a powerful approach to detecting data that falls outside of a model’s training distribution.
Ride Rivals: AI Learns to Compete in Autonomous Vehicle Fleets
New research explores how artificial intelligence can optimize pricing and vehicle distribution when multiple companies operate competing on-demand mobility services.
Beyond the Words: Improving Hate Speech Detection with AI
A new study explores how to refine artificial intelligence models to better identify and combat online hate speech, addressing challenges like limited data and nuanced language.