FAiT: Frequency-Aware Inverted Transformer for Multivariate Time Series Forecasting
Researchers are developing new methods for time series forecasting, focusing on improving accuracy and robustness. Several papers introduce novel attention mechanisms and model architectures designed to better capture complex dependencies, including positive and negative relationships, and to handle non-stationarity and limited data. New benchmarks and evaluation frameworks are also being proposed to rigorously assess these advancements and identify specific failure modes in financial and general time series forecasting. AI
IMPACT Advances in time series forecasting models and benchmarks will improve predictive accuracy and robustness across various domains, including finance and operations.
- cryptocurrency markets
- Diffusion-Copula
- Classification-Diffusion Copula
- Mixture Density Networks
- Physiome-ODE
- Christian Klötergens
- Hugging Face
- arXiv
- Parametric Prior Mapping (PPM)
- PostTime
- L2D-SLDS
- ObsCast
- SiGMA
- Unicorn
- TIPS
- Horizon Activation Mapping
- FinStressTS
- Transformer
- TIME
- Signed Dual Attention
- Time-R1
- SOCK
- SARAF
- TimeOmni-VL
- DAD4TS
- OpenAI