作者位次 #3/4ACL 2026Poster· Natural Language Processing
Anchoring the Affective Manifold: Learning Canonical and Disentangled Representations via Generative Cross-Modal Alignment
Weibin Li · Jintao Cheng · Xiaoyu Tang · Chi-Man VONG
摘要
Dominant multimodal emotion recognition paradigms often neglect the intrinsic geometric structure of affect, resulting in representations heavily entangled with non-affective factors. To address this, we propose a Canonical Disentangled Multimodal Generative Framework aimed at recovering the cano...