BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Information Systems Group - ECPv6.4.0.1//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:Information Systems Group
X-ORIGINAL-URL:https://isg.ics.uci.edu
X-WR-CALDESC:Events for Information Systems Group
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:America/Los_Angeles
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:20260308T100000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:20261101T090000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260424T130000
DTEND;TZID=America/Los_Angeles:20260424T140000
DTSTAMP:20260520T075042
CREATED:20260421T003909Z
LAST-MODIFIED:20260421T060244Z
UID:2422-1777035600-1777039200@isg.ics.uci.edu
SUMMARY:Prof. Eduard Dragut (Temple University): Toward Scalable Knowledge Extraction with Weak Supervision and Large Language Models
DESCRIPTION:Friday April 24\, 2026\, 1:00 PM – 2:00 PM\nDonald Bren Hall 3011\, ICS\, UC Irvine \nZoom:\nhttps://uci.zoom.us/j/95509222811?pwd=2V8Hnx71iP6dyfNsEPoo97NUfCFWTo.1 \nLunch will be provided. \nTitle: Toward Scalable Knowledge Extraction with Weak Supervision and Large Language Models \nAbstract: Information extraction is a foundational capability for transforming unstructured text into structured knowledge\, enabling downstream applications such as knowledge graph construction\, semantic search\, question answering\, and scientific discovery. However\, building high-quality extraction systems traditionally depends on large manually annotated datasets\, which are costly to create and often impractical in specialized domains. In this talk\, I will present recent advances toward scalable information extraction under limited supervision. I will discuss methods for improving the quality of weakly supervised training data through automatic label cleaning\, show how richer benchmarks over full scientific documents expose new challenges for scientific information extraction beyond simplified abstract-level settings\, and demonstrate how large language models can be leveraged in many-shot in-context learning regimes to perform competitive named entity recognition and generate high-quality annotations for low-resource domains. Together\, these results suggest a promising path toward scalable knowledge extraction pipelines that reduce reliance on expensive manual annotation while improving the robustness and adaptability of systems used to build next-generation knowledge graphs and AI applications. \nBio: Eduard Dragut is a Professor in the Department of Computer and Information Sciences at Temple University. He is a senior member of the IEEE. He received his Ph.D. in Computer Science from the University of Illinois at Chicago. His research focuses on data management\, information retrieval\, and applied artificial intelligence\, with an emphasis on building scalable systems for extracting and integrating knowledge from large and heterogeneous data sources. He also pursues interdisciplinary AI projects for social good\, including work on assistive technologies such as augmentative and alternative communication (AAC) and AI-driven tools for knowledge discovery. He has published widely in leading venues in databases\, natural language processing\, and data mining.
URL:https://isg.ics.uci.edu/event/prof-eduard-dragut-temple-university-toward-scalable-knowledge-extraction-with-weak-supervision-and-large-language-models/
LOCATION:DBH 3011
END:VEVENT
END:VCALENDAR