Exploring Advanced Information Extraction: A Survey on Structured Output Generation from Textual Data

PhD Qualifying Examination


Title: "Exploring Advanced Information Extraction: A Survey on Structured 
Output Generation from Textual Data"

by

Mr. Zheye DENG


Abstract:

The rapid increase in textual data has driven the need for effective methods 
to extract and organize information. Transforming unstructured text into 
structured outputs, such as tables, mind maps, and knowledge graphs, can not 
only boost the efficiency of information retrieval for users but also 
facilitate data analysis and visualization, benefitting many downstream 
tasks, including text summarization and text mining. Despite the emergence of 
various benchmarks and methods for this task with the advent of Large 
Language Models (LLMs), a research gap exists in comprehensively summarizing 
and analyzing these methods, datasets, and evaluation metrics. This survey 
aims to bridge this gap by exploring existing techniques for generating 
structured output from text, systematically reviewing the current progress in 
this task, the challenges encountered, and outlining potential directions for 
future research.


Date:                   Wednesday, 26 June 2024

Time:                   4:00pm - 6:00pm

Venue:                  Room 5506
                        Lifts 25/26

Committee Members:      Dr. Yangqiu Song (Supervisor)
                        Prof. Raymond Wong (Chairperson)
                        Dr. Junxian He
                        Prof. Ke Yi