AI Paper Atlas

Research map / Time series / Foundation models for time series

ORTCL: Towards Continual Learning of Time Series Foundation Models on Streaming Data via Orthogonal Rotation

Li Lin, Xinrui Zhang, Qi Zhang, Shuai Wang, Kaiwen Xia

AAAI 2026 · Time series › Foundation models for time series · approach: Time series foundation models · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
apply ORTCL orthogonal rotation updates instead of AlphaEditTimer · ECL single-domain D1MSE: 0.1423 -> 0.1332Improved
apply ORTCL instead of best baselinesTimer · Single-domain D1, 4 datasets avgMSE: avg reduction 2.4% vs best baselineImproved
apply ORTCL instead of best baselinesTimer · Single-domain D4, 4 datasets avgMSE: avg reduction 2.8% vs best baselineImproved
sequentially update with ORTCL instead of full fine-tuning (cross-domain forgetting)Timer · ECL cross-domain, evaluated after streamingMSE: FFT 0.1400 -> 0.1628 vs ORTCL 0.1394 -> 0.1454Improved
sequentially update with ORTCL instead of FFTTimer · Weather cross-domainMAE: FFT-fω 0.2961 vs ORTCL-fω 0.2169Improved
add output-side rotation R (compare w/o R) (ablation)ORTCL · ECL and TrafficMAE: w/o R MAE +11.30% ECL, +11.22% TrafficImproved
add input-side rotation S (w/o S ablation) (ablation)ORTCL · ECL and TrafficMSE/MAEMixed
increase truncation rank k from 64 to 1024 (ablation)ORTCL · ECL and Traffic avg D1-D4MSE/MAENo clear effect
ORTCL vs full fine-tuning per segmentTimer · Single-domain D1-D3MSE/MAEImproved
use AlphaEdit parameter-modifying editing instead of ORTCLTimer · ECL/Traffic D1 hidden representationsrepresentation driftHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy