Preprint
Breaking Shortcut Learning for Cross-Trial EEG-Guided Target Speech Extraction via Two-Stage Training
ArXiv.org
Cornell University
06/23/2026
DOI: 10.48550/arxiv.2606.24164
Abstract
Recent end-to-end models for EEG-guided target speech extraction report impressive results, underscoring potential for neuro-steered hearing technologies. However, our analysis reveals that high within-trial performance can be driven by trial-specific EEG structure that acts as shortcuts for target selection, leading to poor generalization on unseen trials. To overcome this gap, we propose TRUST-TSE, a two-stage framework to mitigate shortcut learning. By introducing contrastive pretraining with attended-speaker negative sampling, we encourage the EEG encoder to capture fine-grained EEG–speech alignment while suppressing trial-identity cues. We also employ a confidence-weighted extraction objective based on EEG–source similarity to guide extraction using the learned representations. Experiments on KUL and DTU datasets show that TRUST-TSE outperforms end-to-end baselines under strict cross-trial protocols, addressing a key reliability bottleneck of existing approaches.
Details
- Title: Subtitle
- Breaking Shortcut Learning for Cross-Trial EEG-Guided Target Speech Extraction via Two-Stage Training
- Creators
- Wonchul ShinInyong ChoiKyogu Lee
- Resource Type
- Preprint
- Publication Details
- ArXiv.org
- DOI
- 10.48550/arxiv.2606.24164
- ISSN
- 2331-8422
- Publisher
- Cornell University; Ithaca, New York
- Language
- English
- Date posted
- 06/23/2026
- Academic Unit
- Communication Sciences and Disorders; Iowa Neuroscience Institute
- Record Identifier
- 9985176656902771
Metrics
1 Record Views