AI Paper Atlas

Research map / Time series / Long-term forecasting

StretchTime: Adaptive Time Series Forecasting via Symplectic Attention

Yubin Kim, Viresh Pati, Jevon Twitty, Vinh Pham, Shihao Yang, Jiecheng Lu

ICML 2026 · Time series › Long-term forecasting · approach: Attention for time series forecasting · 0 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 RoPE with SyPE and adaptive warpStretchTime Transformer · Synthetic warped AR(1), horizon 720MSE: 0.411 -> 0.331Improved
add channel-mixing MLPs (ablation)StretchTime on synthetic warped data · Synthetic warped AR(1), avgMSE: 0.176 -> 0.144Improved
replace RoPE with SyPE plus warp in the Transformer3-layer Transformer · SolarMSE: 0.214 -> 0.195Improved
replace RoPE with SyPE plus warp in the Transformer3-layer Transformer · PEMS03MSE: 0.107 -> 0.094Improved
add learnable symplectic geometry to fixed RoPE with warp (V4 to V5) (ablation)StretchTime · PEMS08 avgMSE: 0.125 -> 0.118Improved
add learnable symplectic geometry to fixed RoPE with warp (V4 to V5) (ablation)StretchTime · PEMS04 avgMSE: 0.090 -> 0.089Improved
replace fixed RoPE with static-time SyPE (V2 to V3) (ablation)Transformer without warp · PEMS04 avgMSE: 0.091 -> 0.091No clear effect
replace fixed RoPE with static-time SyPE (V2 to V3) (ablation)Transformer without warp · PEMS08 avgMSE: 0.129 -> 0.125Improved
add adaptive warp module without positional geometry (V1) vs fixed RoPE (V2) (ablation)Transformer · PEMS04 avgMSE: 0.091 -> 0.093Hurt
add warp to fixed RoPE (V2 to V4) (ablation)Transformer with RoPE · PEMS04 avgMSE: 0.091 -> 0.090Improved
use StretchTime in place of 4 other deep baselines on large/low-noise setsStretchTime · ETTh2 avgMSE: 0.384 vs TimeMixer++ 0.339Hurt
StretchTime vs OLinear on ECLStretchTime · ECL avgMSE: 0.174 vs OLinear 0.159Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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