Guohua Liu

HAFT: Hierarchical Audio-Enhanced Fusion Transformer for Efficient Multimodal Sentiment Analysis

HAFT routes audio–visual interaction through bottleneck tokens with adaptive depth-wise gating before integrating text, jointly reducing attention cost and counteracting modality bias; the Cascaded Audio Feature Enhancement (CAFE) framework strengthens prosodic representations via multi-scale time–frequency extraction; and grouped projections, decoupled positional attention, and layer-wise parameter sharing compress the remaining overhead.