On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,695 papers · 148 categories

Trend · papers per month

count % of corpus

12 results for “iTransformer”

Hedge Fund Portfolio Construction Using PolyModel Theory and iTransformerq-fin.PM

PolyModel theory and iTransformer improve hedge fund portfolio construction.

problem Sparse financial time series data makes portfolio construction challenging.

method Identify asset pool, select risk factors, create quantitative and classical measures, and use iTransformer for trend capture.

result Improved Sharpe ratio and annualized return compared to benchmarks.

Hedge funds Portfolio construction PolyModel theory iTransformer Machine learning

2024-08-06abs ↗pdf ↗

Parking Availability Prediction via Fusing Multi-Source Data with A Self-Supervised Learning Enhanced Spatio-Temporal Inverted Transformercs.LG

This study predicts parking availability using multi-source data and a self-supervised learning enhanced transformer.

problem Accurate parking availability prediction to support urban planning and management.

method Proposes SST-iTransformer, a self-supervised learning enhanced spatio-temporal inverted transformer, integrating multi-source data.

result SST-iTransformer achieves state-of-the-art performance in parking availability prediction.

Parking Availability Prediction Multi-Source Data Self-Supervised Learning Transformer Spatio-Temporal

2025-09-04abs ↗pdf ↗

AverageTime: Enhance Long-Term Time Series Forecasting with Simple Averagingcs.LG

AverageTime uses simple averaging to enhance long-term time series forecasting.

problem Long-term time series forecasting with improved intra-sequence and cross-channel dependencies.

method Proposes AverageTime, a simple, efficient, and scalable forecasting model that reframes channel extraction as a stackable architecture.

result AverageTime surpasses state-of-the-art models in forecasting performance with near-linear complexity.

time series forecasting channel extraction simple averaging sequence information multivariate time series

2024-12-30abs ↗pdf ↗

MMformer with Adaptive Transferable Attention: Advancing Multivariate Time Series Forecasting for Environmental Applicationsstat.AP

MMformer improves forecasting of environmental time series data.

problem Accurately forecasting environmental change trends for policy-making.

method Meta-learning MTS model combining self-attention and adaptive transferable multi-head attention.

result MMformer outperforms other models in air quality and climate datasets, reducing prediction errors by 50% in MSE and 20% in MAE.

Environmental forecasting Meta-learning Time series analysis Air quality Climate change

2025-04-18abs ↗pdf ↗

Pretrained Time-Series Foundation Models for Financial Return Forecastingq-fin.MF

Pretrained time-series models outperform train-from-scratch baselines in financial return forecasting.

problem Financial return forecasting

method Pretrained time-series foundation models

result Pretrained TSFMs dominate the ranking distribution, accounting for 8 of 10 task-level wins.

financial return forecasting pretrained models time-series foundation models TSFMs financial markets

2026-06-25abs ↗pdf ↗

TimeCNN: Refining Cross-Variable Interaction on Time Point for Time Series Forecastingcs.LG

TimeCNN improves forecasting by refining cross-variable interactions over time.

problem Multivariate time series forecasting struggles with dynamic and multifaceted cross-variable correlations.

method TimeCNN uses timepoint-independent convolution kernels to capture evolving relationships among variables.

result TimeCNN outperforms state-of-the-art models in real-world datasets with significant computational and speed advantages.

time series forecasting cross-variable interaction convolutional neural networks multivariate time series dynamic correlations

2024-10-07abs ↗pdf ↗

DecompKAN: Decomposed Patch-KAN for Long-Term Time Series Forecastingcs.LG

DecompKAN improves time series forecasting accuracy and transparency.

problem Accurate and transparent time series forecasting in scientific domains.

method Combines decomposition, patching, normalization, and B-spline KAN edge functions.

result Achieves best or tied-best MSE on 20 of 36 comparisons across 9 datasets.

time series forecasting decomposition patching B-spline KAN transparency

2026-04-27abs ↗pdf ↗

Machine Learning-Based Bitcoin Trading Under Transaction Costs: Evidence From Walk-Forward Forecastingq-fin.TR

Machine learning predicts Bitcoin returns but trading performance drops with costs.

problem Trading Bitcoin predictions with transaction costs.

method XGBoost, LSTM, iTransformer models evaluated in walk-forward protocol; cost-aware execution filter implemented.

result Cost-aware execution filter restores profitability; XGBoost strategy outperforms.

Bitcoin trading machine learning transaction costs walk-forward forecasting XGBoost

2026-05-19abs ↗pdf ↗

A Controlled Comparison of Deep Learning Architectures for Multi-Horizon Financial Forecasting: Evidence from 918 Experimentsq-fin.ST

Study compares nine deep learning architectures for multi-horizon financial forecasting.

problem Evaluating the performance of deep learning architectures for multi-horizon financial forecasting.

method Conducted 918 experiments across cryptocurrency, forex, and equity markets using nine architectures.

result ModernTCN achieves the best mean rank (1.333) with a 75 percent first-place rate.

deep learning financial forecasting multi-horizon cryptocurrency forex

2026-02-27abs ↗pdf ↗

FinStressTS: A Parametric Synthetic Benchmark for Time-Series Forecasting in Financeq-fin.CP

FinStressTS creates synthetic benchmarks for financial forecasting, revealing model weaknesses.

problem Limited failure attribution in real-world financial benchmarks.

method Synthetic benchmark with 30 diagnostic environments linked to six mechanism families.

result Model performance varies by mechanism type, with autoregressive models often outperforming Transformers.

financial forecasting synthetic benchmark model evaluation mechanism-aware deep learning

2026-06-02abs ↗pdf ↗

RG-TTA: Regime-Guided Meta-Control for Test-Time Adaptation in Streaming Time Seriescs.LG

RG-TTA adapts neural forecasters to streaming time series shifts by modulating adaptation intensity.

problem Adapting neural forecasters to distribution shifts in streaming time series data.

method RG-TTA uses a meta-controller that continuously modulates adaptation intensity based on distributional similarity.

result RG-TTA achieves the lowest MSE in 156 of 224 seed-averaged experiments, reducing MSE by 5.7% vs TTA.

Test-Time Adaptation Streaming Time Series Meta-Control Regime-Guided Early Stopping

2026-03-29abs ↗pdf ↗

Heads, Not Backbones: Output Heads Dominate Architectures on Fat-Tailed Returnscs.LG

Deep forecasting models show output heads significantly improve performance on fat-tailed financial returns.

problem Improving deep learning models for forecasting fat-tailed financial returns.

method Comparison of backbone architectures and output heads (point, Gaussian, Gaussian mixture) on S&P 500 monthly log-returns.

result Switching from point to Gaussian heads improves CRPS by about 1.3 percent, and from Gaussian to mixture adds another 2.4 percent.

deep learning financial forecasting fat-tailed returns CRPS Gaussian mixture

2026-06-29abs ↗pdf ↗