AI Paper Atlas

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

Robust Multi-View Fusion via Prototype-Anchored Unbalanced Optimal Transport

Zhang Han, Xingwen Zhao, HUI LI

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
use UOT fusion instead of cross-attention fusion (ablation)UOT-Fusion · NTU 60 X-Sub, 40% missingTop-1 accuracy (%): 83.0 -> 90.6Improved
use unbalanced OT (tau=0.5) instead of near-balanced OT (tau=10) (ablation)UOT-Fusion · NTU 60 X-Sub, 40% missingTop-1 accuracy (%): 87.0 -> 90.6Improved
add shared prototype support (ablation)UOT-Fusion · NTU 60 X-Sub, 40% missingTop-1 accuracy (%): 87.4 -> 90.6Improved
add mass-dependent modality weights lambda (ablation)UOT-Fusion · NTU 60 X-Sub, 40% missingTop-1 accuracy (%): 89.3 -> 90.6Improved
increase atoms and prototypes from J=K=16 to 32 (ablation)UOT-Fusion · NTU 60 X-Sub, 40% missingTop-1 accuracy (%): 90.6 -> 90.9No clear effect
set tau very small (0.05) (ablation)UOT-Fusion · NTU 60 X-Sub, 40% missingTop-1 accuracy (%): 90.6 -> 85.7Hurt
set tau large (2.0) (ablation)UOT-Fusion · NTU 60 X-Sub, 40% missingTop-1 accuracy (%): 90.6 -> 88.1Hurt
use UOT fusion instead of RCMCL in the clean settingUOT-Fusion · NTU 60 X-Sub, 0% missingTop-1 accuracy (%): 96.2 -> 96.0No clear effect
use UOT fusion instead of RCMCL under 80% missingUOT-Fusion · NTU 60 X-Sub, 80% missingTop-1 accuracy (%): 70.8 -> 78.3Improved
replace cosine distance cost with normalized squared Euclidean (ablation)UOT-Fusion · NTU 60Top-1 accuracy (%): ±0.3% accuracyNo clear effect
use UOT fusion under RGB latent noise sigma=2.0UOT-Fusion · NTU 60 X-Sub, RGB noiseTop-1 accuracy (%): 83.1 -> 89.7Improved

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