New frameworks enhance time series forecasting with LLMs and generative models · 6 sources tracked
ByPulseAugur Editorial·[9 sources]·
Researchers are developing advanced frameworks for time series forecasting that integrate diverse data types and provide actionable insights. TokenCast uses LLMs to convert numerical sequences and contextual features into a unified symbolic representation for more accurate predictions. ConTex reformulates counterfactual generation for time series, enabling consistent and real-time interventions by learning a global strategy rather than instance-wise optimization. TimeMoDE addresses data scarcity by using Diffusion Transformers with Mixture-of-Experts, pre-trained on multi-domain datasets. Another approach proposes a flexible framework for modeling predictive distributions of nonlinear time series using conditional generative adversarial networks.
AI
IMPACT
These advancements in time series forecasting could lead to more accurate predictions in critical sectors like finance and healthcare, and enable more actionable insights for decision-making.
RANK_REASON
The cluster contains multiple research papers detailing new methods for time series forecasting.
arXiv:2606.20010v1 Announce Type: new Abstract: Current time series forecasting (TSF) research predominantly focuses on scale-homogeneous data, where different time series share similar numerical magnitude ranges. However, in real-world industrial scenarios such as financial prod…
arXiv:2606.19412v1 Announce Type: new Abstract: Time series forecasting leverages historical patterns to predict future values, but traditional methods face challenges when dealing with complex, non-stationary patterns that are difficult to memorize during training. Retrieval-aug…
Current time series forecasting (TSF) research predominantly focuses on scale-homogeneous data, where different time series share similar numerical magnitude ranges. However, in real-world industrial scenarios such as financial product sales, different time series often differ by…
arXiv:2508.09191v2 Announce Type: replace-cross Abstract: Time series forecasting plays a vital role in supporting decision-making across a wide range of critical applications, including energy, healthcare, and finance. Despite recent advances, forecasting accuracy remains limite…
arXiv cs.LG
TIER_1English(EN)·Jan Voets, Hasan Tercan, Tobias Meisen, Sebastian Baum·
arXiv:2606.18049v1 Announce Type: new Abstract: Decision-making with deep learning-based time series forecasting requires not only accurate predictions but also actionable insights. However, current architectures do not inherently provide such information. Specifically, guidance …
Decision-making with deep learning-based time series forecasting requires not only accurate predictions but also actionable insights. However, current architectures do not inherently provide such information. Specifically, guidance is needed on how current conditions must be modi…
arXiv:2606.15172v1 Announce Type: new Abstract: Synthesizing realistic time series with generative models has wide-ranging applications in real-world scenarios. Despite recent progress, most existing methods are trained under the assumption of abundant training data, which substa…
arXiv stat.ML
TIER_1English(EN)·Jordi Llorens-Terrazas, Mika Meitz·
arXiv:2606.16773v1 Announce Type: cross Abstract: We propose a flexible framework for modeling the predictive distributions of nonlinear, possibly multivariate time series. Our approach expresses a general predictive distribution in an appropriate generative representation that i…
We propose a flexible framework for modeling the predictive distributions of nonlinear, possibly multivariate time series. Our approach expresses a general predictive distribution in an appropriate generative representation that is based on a folklore result from measure theoreti…
