The Hidden Order of Markets
![A model iteratively maps price trajectories, estimating the probability distribution of returns using a single realization-demonstrated with Netflix stock (NFLX) data from July 2022 to July 2023-to reveal the evolution of price and its corresponding probability density function [latex] \hat{\rho}\left(r^{\left(k\right)}\right) [/latex].](https://arxiv.org/html/2604.12197v1/fig1.png)
New research reveals how financial factors aren’t pre-defined, but emerge dynamically from the interconnectedness of assets.
Science
Decoding the AI Hype: When Does Enthusiasm Become a Bubble?
New research offers a robust method for pinpointing periods of irrational exuberance in the age of artificial intelligence.
Predictive Maintenance Gets a Boost from Attention-Based AI
A new deep learning framework leverages the power of attention mechanisms to more accurately forecast equipment failure and remaining useful life.
Predicting the Brain’s Next Move: A New Approach to Neural Forecasting
Researchers have developed a novel method for accurately predicting short-term brain activity using fMRI data, offering improved insights into neural dynamics.
The High-Frequency Forecast: How Institutional Money Changes Prediction Markets
New research examines the impact of large-scale investment on the accuracy and fairness of prediction markets, revealing potential benefits alongside uneven distribution of gains.
Beyond Red Teaming: A New Framework for AI Risk
Researchers propose a proactive approach to AI safety, shifting focus from reactive testing to identifying inherent risks within an AI’s reasoning structure.
The Algorithmic Mismatch: When AI Optimizes for the Wrong Things
A growing body of evidence suggests that the societal harms caused by artificial intelligence aren’t necessarily due to malicious intent, but rather a fundamental disconnect between what algorithms are designed to achieve and what genuinely benefits human well-being.
Mapping Resilience: How AI Can Future-Proof Supply Chains
![The study models spatiotemporal risk propagation within semiconductor supply chains using a five-module process-initialization of network topology [latex]G(V,E)[/latex], endogenous attenuation via recovery rate γ, exogenous filtering based on activation threshold τ, spatial aggregation of upstream disturbances, and state transition-to simulate how disruptions exceeding resilience limits can cascade through the system and potentially trigger systemic collapse.](https://arxiv.org/html/2604.11041v1/Semi-Sim.png)
A new framework leverages generative AI to anticipate disruptions and proactively adapt supply chain strategies for long-term stability.
Decoding Solar Storms: A New Prediction Model
Researchers have developed an advanced machine learning system to better forecast the connection between solar flares and coronal mass ejections, crucial events impacting space weather.
Navigating Financial Complexity with AI Agents
Researchers have developed a new method for rigorously evaluating how well large language models can handle multi-step financial tasks by strategically using external tools.