AI Paper Atlas

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

MindMix: A Multimodal Foundation Model for Auditory Perception Decoding via Deep Neural-Acoustic Alignment

RUI LIU, Zhige Chen, Pengshu, Wenlong You, Zhi-An Huang, Jibin Wu, KC Tan

ICLR 2026 · Vision-language and multimodal › Cross-modal learning · approach: Multimodal fusion with missing modalities · 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
add CALRA cross-modal alignment instead of CLIP-style projection (ablation)MindMix · KULbalanced accuracy: 0.9535 -> 0.9982Improved
use CALRA instead of concat-MLP fusion (ablation)MindMix · HR-EEG4EMObalanced accuracy: 0.8574 -> 0.8878Improved
add bi-directional cross-attention in CALRA (ablation)MindMix · KULbalanced accuracy: 0.9435 -> 0.9982Improved
add shared low-rank alignment in CALRA (ablation)MindMix · KULbalanced accuracy: 0.9742 -> 0.9982Improved
add type-specific aligner in CALRA (ablation)MindMix · KULbalanced accuracy: 0.9853 -> 0.9982Improved
use the proposed pretrained EEG encoder instead of LaBraM/CBraMod/EEGNet (ablation)MindMix · KULbalanced accuracy: 0.9637 (CBraMod) -> 0.9982Improved
use Wav2Vec 2.0 instead of Mel-spectrogram audio features (ablation)MindMix · KULbalanced accuracy: 0.9448 -> 0.9982Improved
use Wav2Vec 2.0 instead of HuBERT (ablation)MindMix · KULbalanced accuracy: 0.9883 -> 0.9982Improved
add audio alignment (multimodal) to EEG encoder (ablation)MindMix · DTU (between-trial)balanced accuracy: 0.6557 -> 0.9543Improved
use trial-disjoint evaluation instead of within-subject random splitDBPNet/DARNet baselines · KULbalanced accuracy: 0.9481 -> 0.6536 (DARNet)Hurt
lengthen decision window from 2s to 5s (ablation)MindMix · HR-EEG4EMObalanced accuracy: 0.8878 -> 0.8917No clear effect
shorten decision window from 2s to 1s (ablation)MindMix · HR-EEG4EMObalanced accuracy: 0.8878 -> 0.8535Hurt

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