Research — page 2
Next-Gen Satellite Data Sharpens Solar Power Forecasting
A new hybrid neural network leverages higher-resolution imagery to improve surface solar irradiance estimates during cloudy weather.
Emergency Text Alerts Can Move the Needle on Energy Demand
High-frequency smart thermostat data shows that statewide emergency alerts successfully drove household energy conservation during a natural gas shortage.
Corporate Social Stances Drive Measurable Revenue Shifts
New transaction data shows that firm stances on social issues create persistent spending gains from aligned customers and losses from opponents.
New Framework for Optimizing Start-up Milestone Strategies
A new stochastic control model provides a mathematical basis for choosing venture activities based on risk-to-cost efficiency.
Information Lags and Social Herding Drive A-Share Volatility
A new agent-based model separates behavioral imitation from information diffusion to explain price overshooting and reversals in China's equity market.
Hybrid AI Pipeline Boosts Product Data Extraction
A new two-step validation method allows mid-size AI models to extract rare product attributes with the precision of much larger systems.
ML-Driven Gas Lift Optimization Boosts Unconventional Yields
A new automated workflow uses Bayesian optimization to increase production in unconventional fields without requiring expensive downhole hardware.
Dynamic Confidence Thresholds for Fail-Degraded AI Perception
A new design strategy allows autonomous systems to maintain safety in poor visibility by adjusting detection sensitivity based on real-time image quality.
New Analytic Method Speeds Up Compound Option Pricing
A Fourier cosine approach eliminates numerical quadrature in multi-stage options, improving efficiency for complex nested decisions.
Counterfactual Data Fixes B2B Propensity Modeling
A new framework replaces linear data augmentation with counterfactual explanations to better predict organizational buying cycles.
Hardware Costs Outweigh Gains for Learned Gating in AMC
New FPGA benchmarks show that adding learned feature gating to modulation classifiers increases logic and latency without improving accuracy.
New Correction Factors Fix Global Inequality Data Gaps
A massive new dataset of 122,351 Gini observations provides the tools to harmonize conflicting economic inequality measures across 222 countries.
New Benchmark Tests AI Speech Accuracy for Earnings Calls
Earnings25 provides a 500-hour dataset to measure how well speech-to-text models handle the specific jargon and structure of financial reporting.
New Outlier-Resistant Tools for Time Series Analysis
Three new estimators provide a computationally efficient way to measure autocorrelation in volatile financial and economic data without outlier distortion.
Short-term carbon price forecasts cut compliance costs
New research shows that optimizing the timing of allowance purchases can lower realized costs for firms in emissions trading systems.
Electricity demand remains rigid despite 90 years of tech
A massive meta-analysis shows that automation and smart metering have failed to make power consumption more responsive to price changes.
New ML Framework Links Blockchain Activity to Market Sentiment
A new classification model integrates on-chain transactions and social media data to decode Bitcoin market emotion with high reliability.
Clustering algorithms outperform manual wind farm data filtering
New research shows automated clustering can identify subtle operational outliers in offshore wind SCADA data more accurately than human experts.
New Generative Framework Solves Angular Data Errors
The ANGLE framework enables precise non-parametric regression for circular data, improving pose estimation and wind prediction.
Small Vision Models Now Rival Frontier AI in Factory Inspection
A new distillation method allows 3B-parameter models to outperform frontier VLMs in industrial defect detection using as few as 18 labeled images.
New AI Module Turns 3D Pose Data Into Biomechanical Insights
BioModule enables standard 3D skeletal tracking to predict how the body loads and activates without requiring new hardware.
RL Agent Optimizes Bitcoin Lightning Network Liquidity
A new graph reinforcement learning approach maximizes routing capacity by strategically placing liquidity under fixed budget constraints.
Stablecoins Fail as Uniform Safe Havens During Market Stress
New registry-based data from Austrian service providers reveals a sharp divide in how retail and institutional actors move assets during financial shocks.
New Framework Fixes Value Forecasting in Recommender Systems
PIT-SUN improves the accuracy of high-value metrics like LTV and GMV by solving a fundamental mathematical conflict in target transformation.
MiLSD Brings Line Detection to Low-Cost Microcontrollers
A new micro-detector enables high-accuracy line segment detection within a sub-megabyte memory budget for edge AI.
TopoBrick Enables Zero-Shot IoT Forecasting via Topology Sampling
A new training-free framework uses building knowledge graphs to select the most relevant exogenous variables for sensor forecasting.
U.S. Economic News Sentiment is Becoming More Persistent
Analysis of 45 years of newspaper data shows sentiment shocks now linger longer and resist short-term correction.
Physics-Based Model Shows Proportional Taxes Fail to Redistribute
New research using the Fokker-Planck framework suggests only progressive taxes can shift wealth distributions within policy-relevant timescales.
New Model Quantifies Coordination Risks in Regime Change
Researchers have developed an estimable framework to separate the feasibility of a coup from the perceived benefits of overthrowing a government.
AdaStop Cuts AI Testing Costs via Dynamic Stopping
A new cost-benefit framework allows developers to stop labeling test data once the marginal value of finding faults drops below the cost of labor.
High-Frequency Data Can Mask Massive Forecasting Errors
New research reveals a 'Granularity Paradox' where finer time-series data improves in-sample fit while compounding out-of-sample failure.
New Sinhala Dataset Enables Fine-Grained Market Sentiment Analysis
The SalAngaBhava dataset allows NLP models to move beyond general sentiment to specific product aspects in a low-resource language.
US Private Ag R&D Shifts Toward State-Level Dispersion
New patent-based data reveals that private agricultural innovation is highly persistent and closely tied to the size of state agricultural economies.
IMF Growth Impact May Be Statistically Negligible
A rigorous reanalysis suggests that the perceived positive effect of IMF programs on economic growth lacks robustness.
High-Res Satellite Data Slashes Farm Boundary Errors
A new dataset proves that 3m resolution imagery drastically improves the mapping of smallholder farms across 24 countries.
New Distillation Framework Cuts Fusion Prediction Costs
A hierarchical knowledge distillation method allows tokamak safety models to maintain multimodal accuracy using only time-series data during inference.
Quantum Koopman Method Unlocks Nonlinear System Simulation
A new data-driven framework allows near-term quantum hardware to simulate complex real-world dynamics by embedding nonlinear systems into linear representations.
Multi-Fidelity Framework Slashes Lattice Design Costs
A new optimization approach reduces computational overhead for material design by 24% while maintaining mechanical performance.
New Algorithm Improves Global Factor Detection in Asset Pricing
An iterative filtering method leverages eigenvector delocalization to identify weak market factors that traditional spectral tests miss.
New Framework Scales AI Audit to Detect Model Shortcuts
ReMoDEx moves image classifier explainability from individual sample inspection to dataset-wide strategy clustering.