Large Language Models (LLMs) for Research Knowledge Discovery
- 7 December 2026
- book chapter
- Published by IGI Global Scientific Publishing
- p. 287-306
Abstract
The growth of the scholarly literature has posed numerous challenges to knowledge discovery and synthesis and research efficiency. Large Language Models (LLMs), founded on the transformer architecture and other state-of-the-art tools like BERT and GPT-type models, have a tremendous potential to revolutionize the way researchers conduct their scholarly work. This chapter evaluates the role that LLMs play in shaping research practices through literature search, evidence screening, knowledge synthesis, hypothesis generation, and academic writing. The chapter offers an evidence-based review of LLMs in the different stages of the research process with the use of the summary tables on model development, application areas, challenges, and examples from various disciplines. Additionally, critical limitations of the current tools will be discussed alongside suggestions for future directions.Keywords
This publication has 15 references indexed in Scilit:
- Transforming Science with Large Language Models: A Survey on AI-assisted Scientific Discovery, Experimentation, Content Generation, and EvaluationACM Computing Surveys, 2026
- Enhancing Quantitative Analysis in Social Sciences With Large Language Models (LLMs): A Methodological Case Study in Educational ResearchAERA Open, 2025
- What is reproducibility in artificial intelligence and machine learning research?AI Magazine, 2025
- Automating Research Synthesis with Domain-Specific Large Language Model Fine-TuningACM Transactions on Knowledge Discovery From Data, 2025
- A Bibliometric Review of Large Language Models Research from 2017 to 2023ACM Transactions on Intelligent Systems and Technology, 2024
- Automating psychological hypothesis generation with AI: when large language models meet causal graphHumanities and Social Sciences Communications, 2024
- Can large language models replace humans in systematic reviews? Evaluating GPT‐4's efficacy in screening and extracting data from peer‐reviewed and grey literature in multiple languagesResearch Synthesis Methods, 2024
- Document-level Relation Extraction via Separate Relation Representation and Logical ReasoningACM Transactions on Information Systems, 2023
- Survey of Hallucination in Natural Language GenerationACM Computing Surveys, 2023
- ChatGPT: five priorities for researchNature, 2023