Key Takeaways
- LLMs can help you synthesise, organise, and identify gaps in your literature review, but they cannot replace reading the papers yourself.
- Never ask an LLM to “find sources” or “cite papers.” LLMs fabricate citations. Use databases (Scopus, Web of Science, Google Scholar) for searching and LLMs for organising what you have already found.
- The best use of LLMs for literature reviews is synthesis: grouping themes, comparing findings, identifying patterns, and improving transitions between paragraphs.
- These 10 prompts cover thematic grouping, synthesis vs summary, identifying gaps, improving transitions, comparing conflicting findings, and structuring the review.
- For a comprehensive guide to using LLMs across all stages of research, see our LLM Prompting Guide for Researchers.
The literature review is often the most time-consuming chapter to write. Not because the reading is hard, but because the synthesis is. You have read 80 papers. You understand them individually. Now you need to weave them into a coherent narrative that builds toward your research gap. That is where most researchers get stuck.
LLMs can help with the synthesis stage, but they have a critical limitation: they hallucinate citations. If you ask an LLM to “find relevant papers on X,” it will generate plausible-looking references that do not exist. Use databases for searching. Use LLMs for organising, comparing, and improving what you have already found.
For a detailed guide to conducting a systematic review, see our post on how to conduct a systematic literature review. For broader LLM guidance, see our LLM Prompting Guide for Researchers.
The 10 Prompts
Prompt 1: Group Sources into Themes
I have the following list of papers for my literature review. Each entry includes the author, year, and a one-sentence summary of the key finding. Group these papers into 3-5 thematic categories based on their topics and findings. For each category, suggest a descriptive heading and list which papers belong to it.
[paste your list of papers with summaries]
Prompt 2: Turn Summaries into Synthesis
The following paragraph from my literature review reads like a list of individual study summaries rather than a synthesis. Rewrite it so that it groups studies by finding, compares and contrasts results, and flows as a coherent narrative. Keep all citations in their current format. Do not add any citations I have not included.
[paste your paragraph]
Prompt 3: Identify Gaps from Summaries
Based on the following summaries of papers in my literature review, identify potential research gaps. What topics are under-explored? What populations or contexts are missing? What methodological approaches have not been tried? What contradictions remain unresolved? Suggest 3-5 potential gaps with a brief justification for each.
[paste your paper summaries]
Prompt 4: Improve Transitions Between Paragraphs
The following two paragraphs from my literature review lack a clear transition. The first paragraph discusses [topic A] and the second discusses [topic B]. Write a transitional sentence that connects them logically and shows how topic B follows from or contrasts with topic A.
Paragraph 1: [paste]
Paragraph 2: [paste]
Prompt 5: Compare Conflicting Findings
The following studies report conflicting findings on [topic]:
Study 1: [author, year, finding]
Study 2: [author, year, finding]
Study 3: [author, year, finding]
Write a paragraph that presents these conflicting findings, suggests possible explanations for the discrepancy (e.g., different methodologies, populations, contexts), and explains why this conflict represents a gap that my study addresses.
Prompt 6: Suggest a Literature Review Structure
My research topic is [topic]. My research questions are: [list questions]. I have reviewed approximately [number] papers. Suggest a logical structure for my literature review with 4-6 section headings. The structure should move from broad context to specific gap, and the final section should set up my research questions. For each section, describe in one sentence what it should cover.
Prompt 7: Summarise a Paper for Your Notes
Summarise the following paper in a structured format for my literature review notes:
1. Full citation
2. Research question/objective (one sentence)
3. Methodology (one sentence)
4. Key findings (two to three sentences)
5. Limitations noted by the authors
6. Relevance to my research on [your topic]
[paste the paper’s abstract or key sections]
Prompt 8: Write the Gap Statement
Based on the literature review content below, write a clear, explicit research gap statement. The statement should: (1) acknowledge what previous studies have established, (2) identify what is missing, understudied, or unresolved, and (3) explain why addressing this gap matters. Keep it to 3-4 sentences.
[paste the final section of your literature review]
Prompt 9: Check for Summary vs Synthesis
Review the following literature review section and identify paragraphs that read as lists of individual study summaries rather than synthesis. For each flagged paragraph, explain what makes it a summary rather than a synthesis, and suggest how to restructure it to compare, contrast, or group findings thematically.
[paste your literature review section]
Prompt 10: Create a Summary Table of Reviewed Studies
Using the following list of papers I have reviewed, create a summary table with these columns: Author(s) & Year, Research Objective, Methodology, Sample Size, Key Findings, Limitations. Format it as a markdown table. I will verify each entry for accuracy before including it in my thesis.
[paste your paper summaries]
Critical Warning: LLMs Fabricate Citations
This cannot be stressed enough. LLMs generate plausible-sounding citations that do not exist. If you ask an LLM to “suggest papers on X,” the titles, authors, and journal names it provides may be entirely fabricated. Never add a citation to your literature review based on an LLM suggestion without first verifying it in an actual database (Scopus, Web of Science, Google Scholar). Use LLMs for organising and synthesising sources you have already found, not for finding new ones.
Need your literature review professionally edited? PM Proofreading’s journal editing service ensures your literature review is well-synthesised, clearly written, and ready for peer review. Upload your manuscript and get a quote.
Further Reading
- LLM Prompting Guide for Researchers: ChatGPT, Claude, and Gemini
- How to Conduct a Systematic Literature Review
- How to Write an Effective Introduction Section
- AI vs Human Proofreaders and Editors
Frequently Asked Questions
Can I use ChatGPT to find sources for my literature review?
No. LLMs fabricate citations. They generate titles, authors, and journal names that sound real but do not exist. Use databases (Scopus, Web of Science, Google Scholar, PubMed) to find sources. Use LLMs only to organise, synthesise, and improve the writing of sources you have already found and verified.
How do I use AI to turn summaries into synthesis?
Paste a paragraph that lists studies one by one and ask the LLM to rewrite it so that studies are grouped by finding, compared and contrasted, and connected with transitional phrases. Use Prompt 2 or Prompt 9 above. Always verify that the rewritten version accurately represents each study’s findings.
Is it ethical to use AI for my literature review?
Using AI to help organise and improve your writing is generally acceptable. Using AI to generate content you present as your own reading and analysis is not. The key distinction is between AI as a writing tool (acceptable) and AI as a research substitute (not acceptable). Check your institution’s AI policy for specific guidelines.
What is the difference between summarising and synthesising in a literature review?
Summarising describes individual studies one by one. Synthesising groups studies by theme, compares their findings, identifies patterns and contradictions, and builds a narrative toward your research gap. Reviewers expect synthesis. If your literature review reads like a list of summaries, use Prompt 9 to identify and fix the problem areas.
