Yung-Hsuan (Franklin) Lai

M.S. Graduate from National Taiwan University | Deep Learning | Computer Vision

franklin.jpg
Taipei, Taiwan
Resume

Hello! I am a research assistant in the Vision and Learning Lab at National Taiwan University, focusing on cross-modality learning such as audio-visual learning and vision-text learning.

My academic journey began in the Communication Engineering Department at National Taiwan University, where I pursued my M.S. degree under the guidance of Professor Yu‑Chiang Frank Wang, specializing in Audio-Visual Learning. For my Master’s thesis, I tackled the challenge of weakly-supervised Audio-Visual Video Parsing. My method involves leveraging large-scale contrastively pre-trained models to generate reliable pseudo labels, which outperformed existing methods in terms of effectiveness. Prior to my postgraduate studies, I obtained a B.S. degree in Electrical Engineering from National Taiwan University.

News

Jul 2, 2024 Our paper Receler and Select and Distill are accepted by ECCV 2024
Jan 16, 2024 Our paper RAPPER is accepted by ICLR 2024
Sep 22, 2023 Our paper VALOR is accepted by NeurIPS 2023
Sep 1, 2023 Graduate with a M.S. degree from National Taiwan University! Start working as a research assistant in the same lab!

Publications

2024

  1. Select_and_Distill.jpg
    Select and Distill: Selective Dual-Teacher Knowledge Transfer for Continual Learning on Vision-Language Models
    Yu-Chu Yu, Chi-Pin Huang, Jr-Jen Chen, Kai-Po Chang, Yung-Hsuan Lai, Fu-En Yang, and Yu-Chiang Frank Wang
    In ECCV, 2024
  2. Receler.jpg
    Receler: Reliable concept erasing of text-to-image diffusion models via lightweight erasers
    Chi-Pin Huang, Kai-Po Chang, Chung-Ting Tsai, Yung-Hsuan Lai, and Yu-Chiang Frank Wang
    In ECCV, 2024
  3. RAPPER.jpg
    RAPPER: Reinforced Rationale-Prompted Paradigm for Natural Language Explanation in Visual Question Answering
    Kai-Po Chang, Chi-Pin Huang, Wei-Yuan Cheng, Fu-En Yang, Chien-Yi Wang, Yung-Hsuan Lai, and Yu-Chiang Frank Wang
    In ICLR, 2024

2023

  1. VALOR.jpg
    Modality-Independent Teachers Meet Weakly-Supervised Audio-Visual Event Parser
    Yung-Hsuan Lai, Yen-Chun Chen, and Yu-Chiang Frank Wang
    In NeurIPS, 2023