AI Paper Atlas

Research map / Architectures and efficiency / Neural architecture search

CARL: Causality-guided Architecture Representation Learning for an Interpretable Performance Predictor

Han Ji, Yuqi Feng, Jiahao Fan, Yanan Sun

ICCV 2025 · Architectures and efficiency › Neural architecture search · approach: Bi-level differentiable NAS · 1 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 disentanglement and intervention losses (lambda1=lambda2=0.5 vs 0) (ablation)CARL · NAS-Bench-201 (0.5% train)Kendall's Tau: 0.601 -> 0.656Improved
add disentanglement loss only (lambda1=0.5, lambda2=0) (ablation)CARL · NAS-Bench-201 (0.5% train)Kendall's Tau: 0.601 -> 0.636Improved
add intervention loss only (lambda2=0.5, lambda1=0) (ablation)CARL · NAS-Bench-201 (0.5% train)Kendall's Tau: 0.601 -> 0.618Improved
raise lambda1 and lambda2 from 0.5 to 1 (ablation)CARL · NAS-Bench-201 (0.5% train)Kendall's Tau: 0.656 -> 0.617Hurt
use GCN encoder instead of GIN or GAT (ablation)CARL · NAS-Bench-201 (0.5% train)Kendall's TauImproved
combine node-level and edge-level disentanglement (ablation)CARL · NAS-Bench-201 (0.5% train)Kendall's TauImproved
use CARL instead of PINAT with 0.5% training dataCARL · NAS-Bench-201Kendall's Tau: 0.549 -> 0.656Improved
use CARL instead of NAR-Former with 0.5% training dataCARL · NAS-Bench-201Kendall's Tau: 0.590 -> 0.656Improved
use CARL with 0.02% training data vs NAR-Former with 0.04%CARL · NAS-Bench-101Kendall's Tau: 0.683 vs 0.651Improved
use CARL in predictor-guided search instead of DCLPCARL · NAS-Bench-101 searchtest accuracy (%): 94.17 vs 94.17No clear effect
use CARL instead of PINATCARL · DARTS CIFAR-10accuracy (%): 97.58 -> 97.67Improved
use CARL instead of WeakNASCARL · TransNAS-Bench-101 Macroaverage rank: 10.49 -> 5.29Improved

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