AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Geometric Embedding Alignment via Curvature Matching in Transfer Learning

Sung Moon Ko, Jaewan Lee, Sumin Lee, Soorin Yim, Sehui Han

ICML 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning · 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 Ricci curvature matching and mapping losses (ablation)GEAR (HV from DK task pair) · hv<-dk validation curvevalidation MSEImproved
add mapping loss alone (ablation)GEAR · hv<-dk validation curvevalidation MSEMixed
GEAR curvature matching versus GATEGEAR · minimum validation lossvalidation loss: 17.5% lower than GATEImproved
replace Ricci curvature matching with Hessian-based regularizer (ablation)GEAR · AS<-ST taskvalidation MSEHurt
tune loss weights over a grid of 1024 combinations (ablation)GEAR · AS->ST and ST->AS validationvalidation RMSE: AS->ST mean 0.4314, selected 0.4033No clear effect
increase latent dimension from 5 to 50 (ablation)GEAR · HV with EF sourcevalidation lossImproved
increase latent dimension beyond 50 (to 200) (ablation)GEAR · HV with EF sourcevalidation lossNo clear effect
vary K-loop iterations from 1 to 5 (ablation)GEAR · validation MSEvalidation MSE: 0.278 -> 0.276 at K=2No clear effect
use GEAR instead of GATE under corrupted labels (HV from CT)GEAR · HV prediction with CT sourcemean RMSE: 1.8080 -> 1.3825Improved
use GEAR instead of GATE under corrupted labels (HV from CCS)GEAR · HV prediction with CCS sourcemean RMSE: 1.3553 -> 0.5656Improved
use GEAR instead of GATE under corrupted labels (CCS from HV)GEAR · CCS prediction with HV sourcemean RMSE: 2.7430 -> 2.6811Improved
use analytic instead of autograd derivatives for curvature (ablation)GEAR · memory for metric derivativesmemory: autograd approximately 85.5x more memory at data size 1Improved

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