AI Paper Atlas

Research map / Vision-language and multimodal / Cross-modal learning

EMBridge: Enhancing Gesture Generalization from EMG Signals Through Cross-modal Representation Learning

Wenhui Cui, Christopher Michael Sandino, Hadi Pouransari, Ran Liu, Juri Minxha, Ellen L. Zippi, Erdrin Azemi, Behrooz Mahasseni

ICLR 2026 · Vision-language and multimodal › Cross-modal learning · approach: Unimodal-to-multimodal representation transfer

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 Q-Former (ablation)EMBridge · emg2pose unseen, zero-shotbalanced accuracy: 0.494 -> 0.528Improved
add Q-Former (ablation)EMBridge · emg2pose in-dist, linear probebalanced accuracy: 0.793 -> 0.785Hurt
add masked pose reconstruction loss (ablation)EMBridge · emg2pose unseen, zero-shotbalanced accuracy: 0.516 -> 0.528Improved
add CASCLe soft contrastive loss (ablation)EMBridge · emg2pose unseen, zero-shotbalanced accuracy: 0.509 -> 0.528Improved
add CASCLe soft contrastive loss (ablation)EMBridge · emg2pose unseen, linear probebalanced accuracy: 0.485 -> 0.505Improved
replace label smoothing with CASCLe (ablation)EMBridge · emg2pose unseen, zero-shotbalanced accuracy: 0.511 -> 0.528Improved
replace SoftCLIP with CASCLe (ablation)EMBridge · emg2pose unseen, zero-shotbalanced accuracy: 0.510 -> 0.528Improved
add label smoothing 0.1 to InfoNCEQ-Former · emg2pose unseen, zero-shotbalanced accuracy: 0.498 -> 0.498No clear effect
increase batch size from 256 to 1024 (ablation)CPEP · emg2pose unseen, zero-shotbalanced accuracy: 0.481 -> 0.433Hurt
increase batch size from 256 to 1024 (ablation)EMBridge · emg2pose unseen, zero-shotbalanced accuracy: 0.528 -> 0.523No clear effect
increase top-k clusters in CASCLe to 10 or 20 (ablation)EMBridge · emg2posebalanced accuracyHurt
use 16 queries instead of 8 or 32 (ablation)EMBridge · emg2posebalanced accuracyImproved

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