aiqbench analysis
This research introduces ORCA, a novel method for adapting Time Series Foundation Models in black-box environments where direct parameter access is unavailable. By leveraging the context of predictive errors rather than relying on gradient backpropagation, the approach enables effective online adaptation across multiple state-of-the-art models and datasets. This work addresses a critical gap as foundation models increasingly shift toward closed-source API services, offering a viable path for continuous learning without white-box access.
Learning the Context of Errors: Black-Box Online Adaptation of Time Series Foundation Models
arXiv:2606.14222v2 Announce Type: replace Abstract: The rapid evolution of Time Series Foundation Models (TSFMs) has advanced zero-shot forecasting across diverse domains. Inspired by the current form of Large Language Models, future TSFMs may be offered as commercialized, closed-source API services. However, many existing online adaptation methods still rely on white-box access for parameter fine-tuning or gradient backpropagation. This paradigm mismatch raises a question: In black-box online adaptation for TSFMs, what should we learn? We answer this with an insight: the predictive errors of the base model are conditioned on both the input and output of the base model (i.e., the context of errors). To validate this insight, we propose ORCA (Online Residual Contextual Adaptation). We conduct extensive experiments across 5 state-of-the-art TSFMs and 8 datasets to demonstrate the effectiveness of our approach. Furthermore, through ablation studies, we quantitatively analyze the impact of different adapter learning hypotheses on the final adaptation performance in black-box online adaptation. Code available at https://github.com/Fifthky/ORCA.
Abstract:The rapid evolution of Time Series Foundation Models (TSFMs) has advanced zero-shot forecasting across diverse domains. Inspired by the current form of Large Language Models, future TSFMs may be offered as commercialized, closed-source API services. However, many existing online adaptation methods still rely on white-box access for parameter fine-tuning or gradient backpropagation. This paradigm mismatch raises a question: In black-box online adaptation for TSFMs, what should we learn? We answer this with an insight: the predictive errors of the base model are conditioned on both the input and output of the base model (i.e., the context of errors). To validate this insight, we propose ORCA (Online Residual Contextual Adaptation). We conduct extensive experiments across 5 state-of-the-art TSFMs and 8 datasets to demonstrate the effectiveness of our approach. Furthermore, through ablation studies, we quantitatively analyze the impact of different adapter learning hypotheses on the final adaptation performance in black-box online adaptation. Code available at this https URL.
Machine Learning (cs.LG)
arXiv:2606.14222 [cs.LG]
(or arXiv:2606.14222v2 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2606.14222 Focus to learn more arXiv-issued DOI via DataCite
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 · 248 words
Related coverage
Learning the Context of Errors: Black-Box Online Adaptation of Time Series Foundation Models
arXiv · Jun 15, 2026
Benchtalks #3: We taught AI everything except how to learn
Snorkel AI Blog · Jun 25, 2026
Runway’s WorldPrompt and the Engineering of Real-Time Worlds
Latent Space · Sep 25, 2026
Continual Learning Bench: measuring whether AI systems actually improve with experience
Snorkel AI Blog · Aug 21, 2026