Beyond Citations: Mapping the Path From Science to Innovation
A new deep learning model reveals semantic connections between scientific publications and patents, offering a richer understanding of knowledge transfer.
Science
Predicting Chaos: A New Approach to Forecasting
![The study introduces a generative framework for forecasting chaotic dynamical systems that models the joint probability distribution of temporal sequences-[latex] p({x}\_{t\_{n}},{x}\_{t\_{1}},{x}\_{t\_{2}},\ldots) [/latex]-enabling forecasts derived through marginalization and intrinsic uncertainty quantification via ensemble variance, autocorrelation, and Wasserstein drift, thereby addressing the challenges posed by high-dimensional chaos and sensitivity to initial conditions.](https://arxiv.org/html/2512.24446v1/figures/vizabs_B.png)
A novel generative modeling framework leverages joint probability distributions to deliver more accurate and statistically consistent long-term predictions for complex, dynamic systems.
Can AI Know Itself? Assessing the Self-Awareness of Large Language Models
A new study explores whether current artificial intelligence systems can accurately gauge their own capabilities before attempting complex tasks.
Beyond the Hotline: How AI Can Prepare, Not Replace, Mental Health Support
New research examines the role of conversational AI in addressing mental health crises, focusing on its potential to facilitate help-seeking behavior and readiness for human intervention.
Predicting Tomorrow’s News with AI
Researchers are leveraging the power of large language models to forecast future events by training them on a massive dataset of real-world questions.
Decoding Sustainability: Can AI Unlock EU Taxonomy Data?
A new study investigates how artificial intelligence can automatically analyze corporate sustainability reports to identify key performance indicators for EU regulatory compliance.
Smarter Rain Predictions for East Africa
A new machine learning approach is delivering more accurate and reliable rainfall forecasts, crucial for communities across East Africa.
When AI Meets Disaster: The Limits of 3D Vision
![Three-dimensional semantic segmentation is achieved by projecting two-dimensional annotations into a volumetric space via majority voting across multiple frames, with subsequent refinement in CloudCompare[cloudcompare] ensuring point-level accuracy.](https://arxiv.org/html/2512.24593v1/Figures/PNG/semantic.png)
New research reveals that current artificial intelligence struggles to accurately assess damage in real-world post-disaster environments.
Mapping Himalayan Glacial Lake Change with AI
A new deep learning approach leverages satellite radar data to automatically monitor glacial lakes in the Himalayas, improving early warning systems for potentially catastrophic outburst floods.
Mapping the Flood: How Graph Networks and Physics Improve Prediction
A new approach combines the power of graph neural networks with fundamental physics principles to deliver more accurate and efficient flood forecasting.