← Back to all news
arXiv Machine Learning By Nouha Karaouli, Denis Coquenet, Elisa Fromont, Martial Mermillod, Marina Reyboz

Foundation vs. Specialized Models: Evaluating Catastrophic Forgetting in Continual Time Series Forecasting

Read the original on arXiv Machine Learning →

arXiv:2510. 00809v3 Announce Type: replace Abstract: While Time Series Foundation Models (TSFMs) excel in zero-shot tasks, their behavior under continual fine tuning is poorly understood.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Related stories

arXiv AI
Jun 16

TS-Memory: Plug-and-Play Memory for Time Series Foundation Models

arXiv:2602. 11550v2 Announce Type: replace-cross Abstract: Time Series Foundation Models (TSFMs) achieve strong zero-shot forecasting through large-scale pre-training, but adapting them to downstream domains under distribution shift remains challenging.

By Sisuo Lyu, Siru Zhong, Tiegang Chen, Weilin Ruan, Qingxiang Liu, Taiqiang Lv, Qingsong Wen, Raymond Chi-Wing Wong, Yuxuan Liang
efficiencybenchmarks
More like this →
arXiv Machine Learning
Jun 25

Towards Continuous Power Forecasting: Practical Continual Learning for Real-World Energy Systems in Nonstationary Time Series

arXiv:2606. 24955v1 Announce Type: new Abstract: Power forecasting models deployed in real-world energy markets must operate under nonstationary conditions, where data distributions continually evolve due to weather variability, infrastructure upgrades, and changing consumption behaviors.

By Yujiang He, Frederic Uhrweiller, Bernhard Sick
More like this →
arXiv AI
Sep 24

Parameter Importance-Driven Continual Learning for Foundation Models

The paper introduces PIECE, a Parameter Importance-Driven Continual Learning method that selectively updates only 0.1% of core parameters to preserve general abilities while learning new domain knowledge. PIECE employs two importance estimators—PIECE‑F using Fisher Information and PIECE‑S combining gradient and curvature information—to guide updates. Experiments on three language models and two multimodal models demonstrate that PIECE maintains general capabilities and achieves state‑of‑the‑art continual learning performance without accessing prior training data or adding parameter overhead.

By Lingxiang Wang, Hainan Zhang, Zhiming Zheng
llmsmultimodalbenchmarks
More like this →