AI Paper Atlas

Research map / Time series / Long-term forecasting

Harmonic Dataset Distillation for Time Series Forecasting

Seungha Hong, Sanghwan Jang, Wonbin Kweon, Suyeon Kim, Gyuseok Lee, Hwanjo Yu

AAAI 2026 · Time series › Long-term forecasting · approach: Attention for time series forecasting · 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
add sinusoidal decomposition (frequency-domain updates) to gradient matching (ablation)HDT, DLinear · ETTh1 (M=384, avg over evaluation models)MSE: 0.583 -> 0.545Improved
add Harmonic Matching on top of decomposition (ablation)HDT, DLinear · ETTh1 (M=384, avg over evaluation models)MSE: 0.545 -> 0.420Improved
add Harmonic Matching on top of decomposition (ablation)HDT, DLinear · Traffic (M=384, avg over evaluation models)MSE: 0.902 -> 0.760Improved
add Harmonic Matching on top of decomposition (ablation)HDT, DLinear · Electricity (M=384, avg over evaluation models)MSE: 0.376 -> 0.226Improved
add sinusoidal decomposition to gradient matching (ablation)HDT, DLinear · ETTm2 (M=384, avg over evaluation models)MSE: 0.402 -> 0.325Improved
use HDT instead of CondTSF with xPatch backbone evaluated on DLinear (ablation)distillation method · ETTh1 (M=384)MSE: 1.172 -> 0.412Improved
use HDT instead of CondTSFdistillation method, DLinear backbone · ETTh1 (M=384, evaluated on DLinear)MSE: 0.510 -> 0.430Improved
use HDT instead of CondTSFdistillation method, DLinear backbone · Traffic (M=384, evaluated on DLinear)MSE: 0.716 -> 0.679Improved
use HDT instead of CondTSF on large CA datasetdistillation method · CA (LargeST, M=384)MSE: 197.95 -> 46.63Improved
train iTransformer on distilled data instead of full dataiTransformer · Electricitytraining time (s): 1650.33 -> 1.98Improved
fine-tune Moirai-Large on HDT few-shot data instead of full fine-tuningMoirai-Large · ETTh1MSE: 1.383 -> 1.417Mixed
use HDT instead of CondTSFdistillation runtime, DLinear · ETTh1 (100 outer loops)runtime (s): 32.13 -> 36.37Hurt

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