AI Paper Atlas

Research map / Time series / Causal and generative time series

Parameter Decorrelation via Transition-Variance Alignment for Multivariate Time-series Forecasting

Ji-Eun Choi, Jae-Hong Lee, Joon-Hyuk Chang

ICML 2026 · Time series › Causal and generative time series · approach: Probabilistic and conformal forecasting

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 TVA gating of the step sizeMLP · SolarMSE: 0.277 -> 0.234Improved
add TVA gating of the step sizeMLP · TrafficMSE: 0.508 -> 0.452Improved
add TVA gating of the step sizeMLP · ETTh2MSE: 0.374 -> 0.371Improved
add TVAMLP · Electricity (T=720)MSE: 0.274 -> 0.253Improved
add TVAiTransformer · Electricity (T=720)MSE: 0.275 -> 0.244Improved
add TVA to normalization methods (RevIN, DishTS, SAN)iTransformer, DLinear, MLP · 8 datasetsMSE: Electricity DishTS iTransformer 0.214 -> 0.247; Solar RevIN iTransformer 0.298 -> 0.276Mixed
add TVA to LinearLinear · SolarMSE: 0.343 -> 0.263Improved
add TVA to DLinearDLinear · ETTh1MSE: 0.446 -> 0.446No clear effect
use TVA instead of cosine annealing scheduler (ablation)MLP · SolarMSE: 0.275 -> 0.234Improved
use TVA instead of ReduceLROnPlateau (ablation)MLP · TrafficMSE: 0.511 -> 0.452Improved
use diagonal covariance instead of mean-field scale in TVA (ablation)MLP+TVA · SolarMSE: 0.234 -> 0.224Improved
use diagonal covariance instead of mean-field scale in TVA (ablation)MLP+TVA · ETTh1MSE: 0.438 -> 0.439No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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