AI Paper Atlas

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

Beyond Magnitude: Scale-Invariant Evidential Fusion for Multi-View Classification

Wei Liu, Yufei Chen, Jie Shi, Xiaodong Yue

ICML 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Uncertainty-aware multi-view learning

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 instance-wise standardization of logits (ablation)SAEF · AVE (semantic hijacking)conflict detection AUROC: 14.06 -> 93.55Improved
use Stouffer fusion instead of Sum or Mean (ablation)SAEF · AVE (sensor noise)noise rejection AUROC: 27.21 (Sum) / 28.19 (Mean) -> 87.03Improved
use Stouffer fusion instead of Max (ablation)SAEF · AVEclean accuracy: 80.10 -> 82.84Improved
instance standardization instead of batch norm (ablation)SAEF · AVEconflict detection AUROC: 88.19 -> 93.55Improved
Stouffer fusion instead of Sum/Mean on clean accuracy (ablation)SAEF · AVEclean accuracy: 83.57 (Sum), 83.66 (Mean) vs 82.84No clear effect
vary canonical scale beta from 1 to 20 (ablation)SAEF · AVEclean accuracy: 83.66 -> 82.41No clear effect
vary noise floor epsilon up to 0.1 (ablation)SAEF · AVEclean accuracy: 82.84 for all epsilon up to 0.1No clear effect
replace SAEF with canonical fusion under logit scale perturbation of one viewDS/Sum/Mean/Weighted fusion · SUNRGBDaccuracy: 48% -> 24%Hurt
use SAEF instead of canonical fusionmulti-view fusion · SUNRGBDaccuracy: 46.03 -> 48.17Improved
use SAEF instead of Mean fusionmulti-view fusion · inference runtimems per sample: latency difference bounded within about 0.1msNo 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