Hung-Min Hsu — accepted AI-conference papers

2024

Meta-Diffu$B$: A Contextualized Sequence-to-Sequence Text Diffusion Model with Meta-Exploration

Yunyen Chuang, Hung-Min Hsu, Kevin Lin, Chen-Sheng Gu, Ling Zhen Li, Ray-I Chang +1

NeurIPS 2024poster

The diffusion model, a new generative modeling paradigm, has achieved significant success in generating images, audio, video, and text. It has been adapted for sequence-to-sequence text generation (Seq2Seq) through DiffuSeq, termed the S2S-Diffusion model. Existing S2S-Diffusion models predominantly…

Cited by 0 Source ↗PDF ↗code ↗Scholar ↗

2022

LUNA: Localizing Unfamiliarity Near Acquaintance for Open-Set Long-Tailed Recognition

Jiarui Cai, Yizhou Wang, Hung-Min Hsu, Jenq-Neng Hwang, Kelsey Magrane, Craig S Rose

AAAI 2022technical

The predefined artificially-balanced training classes in object recognition have limited capability in modeling real-world scenarios where objects are imbalanced-distributed with unknown classes. In this paper, we discuss a promising solution to the Open-set Long-Tailed Recognition (OLTR) task utili…

Cited by 14 Source ↗PDF ↗Scholar ↗