Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Investigating Expert-in-the-Loop LLM Discourse Patterns for Ancient Intertextual Analysis

Ray Umphrey, Jesse Roberts, Lindsey Roberts

TL;DR

The expert-in-the-loop methodology presented offers a scalable approach for intertextual research into the complex web of intertextuality within and beyond the biblical corpus.

Abstract

This study explores the potential of large language models (LLMs) for identifying and examining intertextual relationships within biblical, Koine Greek texts. By evaluating the performance of LLMs on various intertextuality scenarios the study demonstrates that these models can detect direct quotations, allusions, and echoes between texts. The LLM's ability to generate novel intertextual observations and connections highlights its potential to uncover new insights. However, the model also struggles with long query passages and the inclusion of false intertextual dependences, emphasizing the importance of expert evaluation. The expert-in-the-loop methodology presented offers a scalable approach for intertextual research into the complex web of intertextuality within and beyond the biblical corpus.

Investigating Expert-in-the-Loop LLM Discourse Patterns for Ancient Intertextual Analysis

TL;DR

The expert-in-the-loop methodology presented offers a scalable approach for intertextual research into the complex web of intertextuality within and beyond the biblical corpus.

Abstract

This study explores the potential of large language models (LLMs) for identifying and examining intertextual relationships within biblical, Koine Greek texts. By evaluating the performance of LLMs on various intertextuality scenarios the study demonstrates that these models can detect direct quotations, allusions, and echoes between texts. The LLM's ability to generate novel intertextual observations and connections highlights its potential to uncover new insights. However, the model also struggles with long query passages and the inclusion of false intertextual dependences, emphasizing the importance of expert evaluation. The expert-in-the-loop methodology presented offers a scalable approach for intertextual research into the complex web of intertextuality within and beyond the biblical corpus.
Paper Structure (19 sections, 2 figures, 3 tables)

This paper contains 19 sections, 2 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Background
  3. LLMs
  4. Related Work
  5. Definitions and Criteria
  6. Methodology
  7. Selected Texts and Scenarios
  8. Experiments
  9. True Positive: Matthew 7 & 1 Clement 13
  10. False Positive: Matthew 7 & 1 Peter 2:4-8
  11. Probable: Sirach 51 & Matthew 11:25-30
  12. Speculative: Romans 3:19-5:11 & Luke 18:9-14
  13. Hypothesis: Matthew 10:5-42, Acts 20:17-35, & Luke 10:2-12
  14. Out-of-Sample: Matthew 7 & Fish and the Tree
  15. Results and Observations
  16. ...and 4 more sections

Figures (2)

  • Figure 1: Example approximated intertextual dependence over Genesis, Paradise Lost milton2005paradise, and Hocus Pocus HocusPocus1993. Future works depend on past related texts with some weight. Subsequent work transforms the combined inherited weighted representations like a neural activation function.
  • Figure 2: Prompt pattern used to identify intertextualities between the corpus (to be searched) and the query (passage of interest).