안녕하세요.

2026년 10월  7일 (수), 중앙대학교 정다흰 교수님을 모시고 세미나를 진행합니다.

지속가능 IT 기술 세미나는 IT 기술의 지속 가능성과 관련한 최신 연구 및 동향을 다루는 강의로, 다양한 전공 분야의 전문가를 초빙하여 깊이 있는 강연을 제공합니다.

관심 있는 학생 여러분의 많은 참여 바랍니다.

[세미나 상세 정보]

■일시: 2026년 10월 7일 (수) 17:00 ~ 17:50

■장소: 제1공학관 23동 23219호

■연사: 정다흰 교수님 (중앙대학교 AI학과)

■강의 제목:  From Efficient LLMs/VLMs to Robust VLA Models

■Abstract

:  In this seminar, I will introduce several recent research topics spanning efficient and robust multimodal AI. The first part focuses on efficient inference for large language models (LLMs) and vision-language models (VLMs), covering quantization for Mamba-based language models and retraining-free, task-agnostic pruning for VLMs. The second part moves toward Vision-Language-Action (VLA) models, examining their architectural diversity and benchmarks, with a particular focus on robustness to linguistic variations in robot instructions. Together, these topics provide an overview of recent advances from efficient model inference to robust multimodal interaction.

■BIO / CV 

: Dahuin Jung is an Assistant Professor in the Department of Artificial Intelligence at Chung-Ang University. Her research focuses on efficient and reliable AI, particularly on efficient inference for LLMs and VLMs and robustness of Vision-Language-Action (VLA) models. Prior to joining Chung-Ang University, she was an Assistant Professor at Soongsil University and a research intern at Qualcomm AI Research and NAVER AI Lab. She received her Ph.D. in Electrical and Computer Engineering from Seoul National University, where she received the Best Doctoral Dissertation Award, and her B.S. in Electrical Engineering from the University of Washington. Her work has been published at major AI conferences including NeurIPS, ICLR, ICML, CVPR, ICCV, and ECCV, with an ICCV Oral and an ICLR Spotlight.

■HOST : 전정훈 교수 (現 정보통신대학장 / 반도체시스템공학과)