Can GPT-4 Support Analysis of Textual Data in Tasks Requiring Highly Specialized Domain Expertise?

Conference proceeding

Jaromir Savelka, Kevin D. Ashley, Morgan A. Gray, Hannes Westermann and Huihui Xu

CEUR workshop proceedings, Vol.3441, pp.1-12

2023

Abstract

annotation guidelines

batch predictions

chain-of-thought prompting

court opinions

generative pre-trained transformers

GPT-4

legal analysis

model brittleness

semantic annotation

1 Record Views

Title: Can GPT-4 Support Analysis of Textual Data in Tasks Requiring Highly Specialized Domain Expertise?
Author/Creator: Jaromir Savelka - Carnegie Mellon University
Kevin D. Ashley - University of Pittsburgh
Morgan A. Gray - University of Pittsburgh
Hannes Westermann - Université de Montréal
Huihui Xu - University of Pittsburgh
Publication Details: CEUR workshop proceedings, Vol.3441, pp.1-12
Grant note: 2040490 / National Eye Institute (http://data.elsevier.com/vocabulary/SciValFunders/100000053) National Eye Institute (http://data.elsevier.com/vocabulary/SciValFunders/100000053)
Academic Unit: School of Law
Language: English
Resource Type: Conference proceeding
Record Identifier: 991015403687603691