Seeing Seizures Coming: AI Learns to Forecast from Video
![The framework leverages a two-stage process-first, a cross-species continual pre-training of the VideoMAE model using a tube masking strategy and [latex]MSE[/latex] loss to learn robust video representations, and second, the transfer of these learned weights to a forecasting model that predicts seizure onset within a defined future window based on encoded states derived from monitoring clips.](https://arxiv.org/html/2603.12887v1/x1.png)
A new study demonstrates that artificial intelligence, initially trained on animal video data, can accurately predict epileptic seizures in humans using only standard video recordings.
Science
Chasing Fraud: An Adaptive System for E-Commerce Protection
A new system, FraudFox, dynamically adjusts to evolving fraud patterns and business needs, providing a more robust defense against online transaction fraud.
When Bots Go Bad: Securing the Next Generation of Autonomous Agents
As AI-powered agents gain greater autonomy, understanding and mitigating their unique security risks is paramount.
Who You Borrow From Matters More Than What You Borrow
New research reveals that a company’s position within the lending network is increasingly influencing credit access, potentially eclipsing traditional financial metrics.
From Chaos to Clarity: AI-Powered Cyber Risk Management
A new framework uses artificial intelligence to automatically connect real-world cyber incidents to known attack patterns and security defenses, improving threat response and risk mitigation.
Taming Chaos: Forecasting Complex Systems with AI
A new approach combines diffusion models and adaptive sensing to improve the long-term predictability of chaotic dynamics on complex geometries.
Turning Forecasts into Fortification: AI Predicts Tropical Cyclone Impacts
A new framework leverages the speed of artificial intelligence to deliver high-resolution risk assessments for critical infrastructure facing tropical cyclones.
Forecasting the Unlikely: Better Predictions for Rare Events
A new approach leverages diagnostic transport maps to refine probabilistic forecasts and improve their reliability when predicting infrequent occurrences.
Mapping Loan Risk Over Time
New research demonstrates how survival analysis can more accurately predict when loans will default, improving risk management under modern accounting standards.
Beyond Species Barriers: Predicting Antibiotic Resistance with Genomic Models
![MiniRocket and Global Pooling exhibit contrasting performance based on phylogenetic distance, with MiniRocket maintaining accuracy across increasing distances on validation sets while Global Pooling excels on unseen test sets, demonstrating that antibiotic resistance mechanisms-rather than phylogenetic relatedness alone-are the primary determinants of predictive power in assessing ampicillin resistance, a finding consistently observed across replicate analyses and further substantiated by metrics such as Matthews correlation coefficient [latex] MCC [/latex].](https://arxiv.org/html/2603.11141v1/figures/ampicillin_v1-1_phylogenetic_distance.png)
New research reveals that the key to accurately forecasting antibiotic resistance across different bacteria lies in how genomic information is analyzed, matching the method to the underlying genetic mechanisms.