Researchers have developed ORCA (Online Residual Contextual Adaptation), a novel method for adapting Time Series Foundation Models (TSFMs) in a black-box setting. This approach focuses on learning from the predictive errors of the base model, recognizing that these errors are conditioned on the model's input and output. The method was validated through extensive experiments on five state-of-the-art TSFMs and eight datasets, demonstrating its effectiveness in improving adaptation performance without requiring white-box access. AI
IMPACT This research offers a new approach for adapting complex time series models when only black-box access is available, potentially broadening their applicability in commercial settings.
RANK_REASON The cluster contains a research paper detailing a new method for adapting AI models.
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
