Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Transforming Scholarly Landscapes: Influence of Large Language Models on Academic Fields beyond Computer Science

Aniket Pramanick, Yufang Hou, Saif M. Mohammad, Iryna Gurevych

TL;DR

This study addresses the problem of understanding how Large Language Models influence fields outside Computer Science. It compiles 106 foundational LLM papers and analyzes roughly 148k non-CS papers that cite LLMs, using Semantic Scholar data from 2018 through February 2024 to quantify adoption, usage patterns, and applications. Key findings show that Linguistics and Engineering lead non-CS LLM citations, BERT remains the most-cited LLM by average yearly citations in non-CS fields, and many disciplines employ task-agnostic models for domain-specific problems, with ethical-risk mentions remaining relatively infrequent (~2.01%). The work provides a detailed, data-driven view of NLP’s cross-disciplinary impact, informing NLP researchers about opportunities for deeper interdisciplinary collaboration, while highlighting the need for domain-tailored models and better risk communication in non-CS contexts.

Abstract

Large Language Models (LLMs) have ushered in a transformative era in Natural Language Processing (NLP), reshaping research and extending NLP's influence to other fields of study. However, there is little to no work examining the degree to which LLMs influence other research fields. This work empirically and systematically examines the influence and use of LLMs in fields beyond NLP. We curate $106$ LLMs and analyze $\sim$$148k$ papers citing LLMs to quantify their influence and reveal trends in their usage patterns. Our analysis reveals not only the increasing prevalence of LLMs in non-CS fields but also the disparities in their usage, with some fields utilizing them more frequently than others since 2018, notably Linguistics and Engineering together accounting for $\sim$$45\%$ of LLM citations. Our findings further indicate that most of these fields predominantly employ task-agnostic LLMs, proficient in zero or few-shot learning without requiring further fine-tuning, to address their domain-specific problems. This study sheds light on the cross-disciplinary impact of NLP through LLMs, providing a better understanding of the opportunities and challenges.

Transforming Scholarly Landscapes: Influence of Large Language Models on Academic Fields beyond Computer Science

TL;DR

This study addresses the problem of understanding how Large Language Models influence fields outside Computer Science. It compiles 106 foundational LLM papers and analyzes roughly 148k non-CS papers that cite LLMs, using Semantic Scholar data from 2018 through February 2024 to quantify adoption, usage patterns, and applications. Key findings show that Linguistics and Engineering lead non-CS LLM citations, BERT remains the most-cited LLM by average yearly citations in non-CS fields, and many disciplines employ task-agnostic models for domain-specific problems, with ethical-risk mentions remaining relatively infrequent (~2.01%). The work provides a detailed, data-driven view of NLP’s cross-disciplinary impact, informing NLP researchers about opportunities for deeper interdisciplinary collaboration, while highlighting the need for domain-tailored models and better risk communication in non-CS contexts.

Abstract

Large Language Models (LLMs) have ushered in a transformative era in Natural Language Processing (NLP), reshaping research and extending NLP's influence to other fields of study. However, there is little to no work examining the degree to which LLMs influence other research fields. This work empirically and systematically examines the influence and use of LLMs in fields beyond NLP. We curate LLMs and analyze papers citing LLMs to quantify their influence and reveal trends in their usage patterns. Our analysis reveals not only the increasing prevalence of LLMs in non-CS fields but also the disparities in their usage, with some fields utilizing them more frequently than others since 2018, notably Linguistics and Engineering together accounting for of LLM citations. Our findings further indicate that most of these fields predominantly employ task-agnostic LLMs, proficient in zero or few-shot learning without requiring further fine-tuning, to address their domain-specific problems. This study sheds light on the cross-disciplinary impact of NLP through LLMs, providing a better understanding of the opportunities and challenges.
Paper Structure (20 sections, 2 equations, 13 figures, 8 tables)

This paper contains 20 sections, 2 equations, 13 figures, 8 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Scientific Trends Analysis.
  4. NLP Scientometrics.
  5. LLM Abilities Assessment.
  6. Data
  7. Analysis
  8. Breadth of LLM Adoption
  9. Evolving Usage Patterns of LLMs
  10. Applications of LLMs in non-CS Fields
  11. Discussion: Opportunities and Challenges
  12. Appendix: Supplementary Definitions and Discussions
  13. Choosing the LLMs
  14. S2 Dataset Statistics and Reliability
  15. Examining LLM citation within CS
  16. ...and 5 more sections

Figures (13)

  • Figure 1: % citations from non-CS fields to LLMs (left); % CS authors collaborating with non-CS fields (right).
  • Figure 2: Inequality in citation distribution across fields.
  • Figure 3: % Papers in non-CS fields (Y-Axis) citing LLMs.
  • Figure 4: Avg. number of CS authors per paper by field (Y-Axis).
  • Figure 5: Number of fields surpassing the threshold number of citations for different NLP Technologies.
  • ...and 8 more figures