BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Department of Electrical and Computer Engineering (HKUECE) 電機與計算機工程系 - ECPv6.15.20//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:Department of Electrical and Computer Engineering (HKUECE) 電機與計算機工程系
X-ORIGINAL-URL:https://ece.hku.hk
X-WR-CALDESC:Events for Department of Electrical and Computer Engineering (HKUECE) 電機與計算機工程系
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:Asia/Hong_Kong
BEGIN:STANDARD
TZOFFSETFROM:+0800
TZOFFSETTO:+0800
TZNAME:HKT
DTSTART:20250101T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20260519T160000
DTEND;TZID=Asia/Hong_Kong:20260519T170000
DTSTAMP:20260511T061102
CREATED:20260420T064633Z
LAST-MODIFIED:20260420T064633Z
UID:115720-1779206400-1779210000@ece.hku.hk
SUMMARY:RPG Seminar – From Understanding to Intervention: Interpretability-Guided Methods for Improving Large Language Models
DESCRIPTION:  \nAbstract\nLarge language models have achieved impressive performance\, but improving them efficiently and reliably requires more than scaling alone. In this talk\, I present a series of works that explore how internal understanding of LLMs can be translated into practical interventions for better capability\, efficiency\, and controllability. I begin with actionable mechanistic interpretability\, introducing a unified “Locate\, Steer\, and Improve” perspective that turns model analysis into a framework for intervention. I then show how this perspective supports several concrete advances: data-free mixed-precision quantization guided by numerical and structural sensitivity\, multilingual capability enhancement through representation shifting and contrastive alignment\, personalized multi-teacher distillation that routes each prompt to its most suitable teacher\, and coarse-to-fine selective fine-tuning for mitigating catastrophic forgetting while preserving general versatility. Together\, these works reflect a common theme: interpretability is not only a tool for explaining LLMs\, but also a principled basis for designing more efficient training\, compression\, and adaptation methods. \nSpeaker\nMr Hengyuan ZHANG\nDepartment of Electrical and Computer Engineering\nThe University of Hong Kong \nBiography of the Speaker\nHengyuan Zhang is a Ph.D. candidate at the University of Hong Kong\, supervised by Prof. Ngai Wong and Prof. Hayden Kwok-Hay So. His research focuses on the improvement of efficiency and interpretability within large language models. He aims to uncover and characterize the internal processes that govern model behavior\, with the goal of improving model speciality\, interpretability\, and reliability in real-world deployments. He has published multiple papers in leading venues such as ACL\, EMNLP\, TKDD\, and NeurIPS. \nOrganiser\nProf. Ngai WONG \nDepartment of Electrical and Computer Engineering\, The University of Hong Kong \nAll are welcome.
URL:https://ece.hku.hk/events/20260519/
LOCATION:Room CB-603\, 6/F\, Chow Yei Ching Building\, The University of Hong Kong
CATEGORIES:Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2024/11/rpg-seminar.jpg
END:VEVENT
END:VCALENDAR