AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

Lost in the Non-convex Loss Landscape: How to Fine-tune the Large Time Series Model?

Xu Zhang, Peng Wang, Wei Wang

ICLR 2026 · Architectures and efficiency › Pruning and efficient tuning · approach: Parameter-efficient transfer analysis · 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
replace full fine-tuning with smoothed full fine-tuning (SFF)Timer · TSF datasets (horizon 96)MSE: average 3% and up to 6.5% MSE reductionImproved
replace full fine-tuning with SFFTimer · ETTh2 (100% data)MSE: 0.3047 -> 0.2737Improved
replace full fine-tuning with SFFTimesFM · TSF datasetsMSE: average improvement of 11.45%Improved
replace full fine-tuning with SFFMOMENT · TSF datasetsMSE: average improvement of 8.31%Improved
replace full fine-tuning with SFFTimer · Anomaly detection Group 1predicted MSE on anomalous segments: 0.072 -> 0.136Improved
replace linear probing then fine-tuning with SFFTimer · TSF Exchange (avg 1-3% data)MSE: 0.4801 -> 0.0856Improved
add weight smoothing before zero-shot inferenceTimer · zero-shot TSF, seven datasetsMSE: average gain of 6.13%Improved
add weight smoothing before zero-shot inferenceTimesFM · zero-shot TSF, seven datasetsMSE: average gain of 35.75%Improved
set interpolation coefficient alpha near 0.85 (ablation)SFF · zero-shot TSF horizon 96MSE: alpha of 0.85 gives best zero-shot performanceImproved
add smoothing to Timer fine-tuning (ablation)Timer · Test MSE convergenceconvergence epochs: both converge within the first epochNo clear effect
replace full fine-tuning with SFFTTMs · ETTm1 (horizon 96)MSE: 0.308 -> 0.309No clear effect
replace full fine-tuning with SFFUniTS · Electricity (horizon 96)MSE: 0.476 -> 0.309Improved

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