AI Paper Atlas

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

Ensemble Prediction of Task Affinity for Efficient Multi-Task Learning

Afiya Ayman, Ayan Mukhopadhyay, Aron Laszka

ICLR 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Dense features and foundation priors

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 residual correction stage (ablation)ETAP · Chemical group gain prediction (|G_train|=10)Pearson correlation: 0.36 -> 0.50Improved
add residual correction stage (ablation)ETAP · CelebA group gain prediction (|G_train|=10)Pearson correlation: 0.31 -> 0.45Improved
add residual correction stage (ablation)ETAP · ETTm1 group gain prediction (|G_train|=10)Pearson correlation: 0.84 -> 0.85No clear effect
use B-spline mapping instead of affine mapping (ablation)ETAP · ETTm1 affinity-to-gainR2: 0.18 -> 0.68Improved
use B-spline mapping instead of affine mapping (ablation)ETAP · CelebA affinity-to-gainR2: 0.13 -> 0.13No clear effect
use ETAP affinity score instead of TAGTAG · Ridership pairwise affinity correlationPearson correlation: 0.10 -> 0.36Improved
use ETAP affinity score instead of TAGTAG · CelebA affinity runtimehours: 6.88 -> 2.51Improved
use ETAP instead of MTGNet with 5 training groupsMTGNet · CelebA group gain predictionPearson correlation: 0.10 -> 0.41Improved
use ETAP instead of Linear SurrogateLinear Surrogate · ETTm1 transfer sign classificationF1: 0.18 -> 0.31Improved
use ETAP task grouping instead of PCGradPCGrad · ETTm1total loss: 4.19 -> 3.89Improved
use ETAP grouping instead of naive MTL with 4 groupsNaive MTL · Chemicaltotal loss: 4.83 -> 4.83No clear effect
apply PCGrad to all tasksNaive MTL · Chemicaltotal loss: 4.83 -> 6.20Hurt

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