AI Paper Atlas

Research map / Graph learning / GNN architectures

Graph Coloring for Multi-Task Learning

Santosh Patapati, Ian Noronha

ECCV 2026 · Graph learning › GNN architectures · approach: Graph structure learning · 2 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 dynamic recoloring instead of static one-shot coloring (ablation)SON-GOKU · CIFAR-10accuracy (%): 61 -> 65Improved
add dynamic recoloring instead of static one-shot coloring (ablation)SON-GOKU · NYUv2depth RMSE: 0.65 -> 0.59Improved
add history-averaged conflict estimates instead of single-step (H=1) (ablation)SON-GOKU · CIFAR-10accuracy (%): 40 -> 65Improved
add history-averaged conflict estimates instead of single-step (H=1) (ablation)SON-GOKU · MM-IMDbaccuracy (%): 20 -> 61Improved
use thresholded conflict graph instead of signed-only graph (ablation)SON-GOKU · CIFAR-10accuracy (%): 56 -> 65Improved
use thresholded conflict graph instead of kNN-symmetric graph (ablation)SON-GOKU · CIFAR-10accuracy (%): 60 -> 65Improved
use quantile threshold graph instead of fixed threshold graph (density-matched) (ablation)SON-GOKU · NYUv2depth RMSE: 0.59 -> 0.59No clear effect
add SON-GOKU scheduling on top of AdaTaskAdaTask · CIFAR-10accuracy (%): 63 -> 67Improved
add SON-GOKU scheduling on top of AdaTaskAdaTask · MM-IMDbaccuracy (%): 59 -> 63Improved
add SON-GOKU scheduling on top of PCGradPCGrad · F&Baccuracy (%): 50 -> 54Improved
add SON-GOKU scheduling on top of PCGradPCGrad · HEALTHaccuracy (%): 58 -> 62Improved
add SON-GOKU scheduling with GradNorm warm startGradNorm · NYUv2surface-normal angle error: 21.4 -> 19.6Improved

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