AI Paper Atlas

Research map / Time series / Foundation models for time series

Residual-Guided Multi-Resolution Refinement of Foundation Models: A Case Study in Drought Forecasting

Wentao Gao, Jiuyong Li, Lin Liu, Thuc Duy Le, Jixue Liu, Yanchang Zhao, Yun Chen

ICML 2026 · Time series › Foundation models for time series · approach: Time series foundation models · 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
add RGMR wrapper to frozen TimesFMTimesFM v1.0 · SPEI Location 1MSE: 0.391 -> 0.318Improved
add RGMR wrapper to frozen TimeGPT-1TimeGPT-1 · SPEI Location 1MSE: 0.437 -> 0.325Improved
add RGMR wrapper to frozen TabPFN-tsTabPFN-ts v1.0 · SPEI Location 1MSE: 0.462 -> 0.344Improved
use frozen foundation model instead of generic deep baselinesTimesFM v1.0 · SPEI Location 1MSE: 0.533 (DLinear) -> 0.391Improved
add coarse projection only (ablation)frozen TimesFM v1.0 · SPEI avg 3 sitesMSE: 0.411 -> 0.387Improved
add multi-resolution proposals without refinement (ablation)frozen TimesFM v1.0 · SPEI avg 3 sitesMSE: 0.387 -> 0.372Improved
add residual-guided updates (ablation)TimesFM multi-resolution · SPEI avg 3 sitesMSE: 0.372 -> 0.341Improved
add adaptive residual weights (ablation)RGMR (TimesFM) · SPEI avg 3 sitesMSE: 0.341 -> 0.334Improved
add adaptive residual weights (ablation)RGMR (TabPFN-ts) · SPEI avg 3 sitesMSE: 0.380 -> 0.367Improved
RGMR refinement at finest stride r=1 (level 5) (ablation)RGMR · South Australia test setsample-wise improvement ratio: 48.5%No clear effect
RGMR refinement from stride 12 to 6 (level 2) (ablation)RGMR · South Australia test setsample-wise improvement ratio: 39.7%Hurt
change resolution scale from [12,6,3,2,1] to alternatives (ablation)RGMR (TimesFM) · SPEI Location 1MSE: 0.318 -> 0.322 to 0.330No clear effect

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