Hsin-Min Wang

Sparse Graph-Guided Mamba is proposed, a lightweight AVSE framework that integrates a sparse heterogeneous graph with a linear-complexity Mamba backbone that explicitly models modality-specific relations through content-adaptive attention and cross-frame audio-visual connections, while Mamba captures long-range temporal context.

CUEES (Correlation-gUided Encoder Selection), a lightweight heuristic that estimates complementarity through task- and category-level Pearson correlations between encoders'performance profiles, scoring a candidate set from single-encoder evaluations alone--without fusion training during selection.

A decoupled optimization framework for text-to-music evaluation significantly outperforms traditional methods by aligning model outputs more closely with human judgments.

Dramatically improve ASR for low-resource languages by cleverly leveraging readily available subtitles in higher-resource languages.