aiqbench analysis
AdaCast introduces a novel framework for time-series foundation models by generating input-specific, low-rank parameter updates to address the limitations of static adaptation methods. This approach enables models to dynamically tailor forecasts to unique temporal patterns and seasonality inherent in heterogeneous datasets, significantly improving zero-shot generalization across domains. The method represents a meaningful advancement in making pre-trained foundation models more flexible for real-world forecasting tasks without requiring full retraining.
AdaCast: Conditional Parameter Generation for Adaptive Time Series Forecasting
arXiv:2610.12240v1 Announce Type: cross Abstract: Time-series foundation models (TSFMs) have achieved strong forecasting performance across domains. However, most adaptation methods remain static. Existing all-in-one methods learn a single set of dataset-level parameter updates and apply the same adapted model to every input. As a result, they cannot adapt the model parameters to the temporal patterns, seasonality and dynamics of each input time series. This limits their ability to produce forecasts that are tailored to heterogeneous inputs. To address this limitation, we propose AdaCast, a conditional parameter generation framework for time-series forecasting. AdaCast uses a generator to produce input-specific low-rank parameter updates for a frozen pretrained TSFM. These updates adapt the model to each input during both training and inference. Across six public benchmarks, AdaCast consistently outperforms static adaptation baseline in in-domain forecasting and improves zero-shot generalization to held-out datasets across domains. These results demonstrate that conditional parameter generation provides an effective approach for adaptive forecasting.
Abstract:Time-series foundation models (TSFMs) have achieved strong forecasting performance across domains. However, most adaptation methods remain static. Existing all-in-one methods learn a single set of dataset-level parameter updates and apply the same adapted model to every input. As a result, they cannot adapt the model parameters to the temporal patterns, seasonality and dynamics of each input time series. This limits their ability to produce forecasts that are tailored to heterogeneous inputs. To address this limitation, we propose AdaCast, a conditional parameter generation framework for time-series forecasting. AdaCast uses a generator to produce input-specific low-rank parameter updates for a frozen pretrained TSFM. These updates adapt the model to each input during both training and inference. Across six public benchmarks, AdaCast consistently outperforms static adaptation baseline in in-domain forecasting and improves zero-shot generalization to held-out datasets across domains. These results demonstrate that conditional parameter generation provides an effective approach for adaptive forecasting.
Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
arXiv:2610.12240 [cs.LG]
(or arXiv:2610.12240v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2610.12240 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 247 words
Related coverage
DiTS: Multimodal Diffusion Transformers Are Time Series Forecasters
arXiv · Oct 8, 2026
NeST: Neighborhood-aware semantic alignment and temporal modulation for LLM based time series forecasting
arXiv · Aug 25, 2026
Introducing Timestral
Mistral AI · Oct 15, 2026
News
Quantum Computing Report · Jul 25, 2026