AI Paper Atlas

Research map / Architectures and efficiency / Mixture of experts

ExPLoRe: Expert Patch-Level Loss Routing for Multi-Objective Masked Image Modeling

Konstantinos Georgiou, Maofeng Tang, Hairong Qi

ECCV 2026 · Architectures and efficiency › Mixture of experts · approach: Sparse MoE as regularization and attention · 0 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
Let loss gradients flow through dispatch weights (loss-coupling) (ablation)ExPLoRe 2-expert · ImageNet-1K k-NN (Token+CLS)k-NN@20 acc: 73.8 -> 75.4Improved
Add dispatch-weight loss weighting (ablation)2-expert MoE · ImageNet-1K k-NN (Token+CLS)k-NN@20 acc: 74.1 -> 75.4Improved
Add dispatch-weight loss weighting (ablation)64-expert MoE · ImageNet-1K k-NN (Token+CLS)k-NN@20 acc: 75.3 -> 76.2Improved
Add dispatch-weight loss weighting (ablation)32-expert MoE · ImageNet-1K k-NN (Token+Pixel)k-NN@20 acc: 73.0 -> 73.7Improved
Add 2-expert MoE without loss weighting (ablation)Token+CLS base · ImageNet-1K k-NNk-NN@20 acc: 75.7 -> 74.1Hurt
Add 2-expert ExPLoRe to Token+CLSToken+CLS base · ImageNet-1K linear probetop-1 acc: 76.2 -> 79.6Improved
Add 2-expert ExPLoRe to Token+CLSToken+CLS base · ImageNet-1K k-NNk-NN@20 acc: 75.7 -> 75.4No clear effect
Remove entropy regularization or set lambda=0.01 (ablation)ExPLoRe 2-expert · ImageNet-1K k-NNk-NN@20 acc: 75.4 -> 66.9 (lambda 0.01); 2.1 (none)Hurt
Use combine-weight loss weighting instead of dispatch weights (ablation)ExPLoRe 2-expert · ImageNet-1K k-NNk-NN@20 acc: text: collapses to 2.1; Table 2 combine reg. 75.4Mixed
Move loss block from 11 to 5/7/9 (ablation)ExPLoRe 2-expert · ImageNet-1K k-NNk-NN@20 acc: 75.4 -> 75.2-75.4No clear effect
Scale Token+Pixel from 32 to 64 expertsExPLoRe Token+Pixel · ImageNet-1K k-NNk-NN@20 acc: 73.7 -> 69.3Hurt
Replace global GradNorm weighting with ExPLoReToken+Pixel · ImageNet-1K k-NNk-NN@20 acc: 32.5 -> 73.7Improved

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