New research details how combining the power of large language models with knowledge graphs can dramatically improve decision-making and accelerate enterprise-level digital transformation initiatives.
Researchers are exploring how artificial intelligence can better understand and identify harmful behaviors within niche online subcultures, offering a path toward more effective intervention.
![A comparison of convergence rates demonstrates that the Standard Kolmogorov metric stagnates due to tail noise, achieving a rate of [latex]n^{-0.25}[/latex], while a Weighted Metric-with a parameter of [latex]q=1.2[/latex]-successfully filters outliers to restore the optimal Gaussian convergence rate of [latex]n^{-0.5}[/latex], thereby accelerating model validation for Student-t distributions ([latex]\nu=2.5[/latex]).](https://arxiv.org/html/2601.04490v1/figs/compare_student.png)
A novel approach to evaluating risk models overcomes limitations with heavy-tailed asset returns, providing more reliable backtesting results.
A new approach combines open-source models to deliver enhanced performance in both Thai language understanding and financial domain expertise.
A new framework systematically exposes domain-specific risks in large language models by crafting subtly harmful prompts.
![FaST demonstrated superior long-horizon forecasting capabilities-predicting [latex]672[/latex] steps into the future based on the preceding [latex]96[/latex]-outperforming both temporal-centric and spatial-temporal-centric methods across sixteen distinct prediction tasks, indicating a substantial advancement in predictive modeling.](https://arxiv.org/html/2601.05174v1/x3.png)
A new framework tackles the challenge of forecasting long-term trends in massive, interconnected networks.
A new machine learning framework enhances the accuracy and reliability of U.S. Treasury yield curve forecasts, even amidst economic turbulence.
A new approach enables dynamic, real-time prediction of critical events from continuous sensor streams, offering improved accuracy and scalability.
A new study reveals how the Truth Social platform amplified unverified claims during the 2024 election, contributing to a volatile information environment.
A new framework reveals how interconnectedness and spatial dynamics amplify the impact of technology in financial systems, offering insights into overcoming adoption barriers.