Author

Hong-Yu Wu

1 paper indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

#artificial intelligence Preprint Sep 2026

SeRV: Semantic-Aligned Residual Vector Quantization for American Sign Language Generation

SeRV (Semantic-Aligned Residual Vector Quantization), a semantic-aligned RVQ tokenizer for ASL generation, achieves state-of-the-art pose accuracy on both How2Sign and YouTube-ASL datasets, while producing semantically consistent 3D ASL motion directly from text.

Hong-Yu Wu, Xu-Ying Wu, Tian-Hao Wu et al. · 0 citations