12,445 papers · continuously updated · last export: 10 Aug 2026livingmeta.ai
← Browse all papers
Research theme

AI Peer Review

The AI Peer Review theme comprises 264 papers in this corpus published between 1981 and 2026. Work here is dominated by Empirical Study, Experimental, Design Science. 5 open research gaps have been surfaced in this area.

Methodology profile

  • Empirical Study45 (17%)
  • Experimental32 (12%)
  • Design Science29 (11%)
  • Position Paper29 (11%)
  • Literature Review26 (10%)
  • Case Study22 (8%)

Research domains

  • Peer Review261 (99%)
  • Academic Writing2 (1%)
  • Research Integrity1 (0%)

Frequent sub-topics

evaluative misalignment in LLM-assisted policy analysis and qualitative coding · 1LLM-based reviewer matching using deep retrieval and deep thinking with information-gain rewards · 1LLM clinical decision support in oncology multidisciplinary tumor boards · 1ChatGPT evaluation of student essays compared to human assessment · 1RAG-based evidence assessment for peer review · 1bias in LLM-assisted peer review · 1agentic AI impact on scientific research and peer review · 1interaction effects between LLM-assisted reviews and LLM-assisted papers · 1

Open research gaps

  • The authors identify several gaps: (1) Limited application beyond computer science domain; (2) Resource constraints limiting exploration of larger model sizes; (3) Need for more sophisticated designs
  • Larger multicenter studies with standardized methods are necessary before widespread adoption of LLMs for automated feedback in medical imaging education can be justified. The authors identify that de
  • Near-total absence of systematic empirical research on prompt injection in academic manuscripts
  • Lack of explicit mention of AI-related misconduct (including prompt injection) in most academic integrity frameworks and misconduct definitions
  • Gap between technology developers and ethical regulators regarding clarity of misconduct definitions

Representative papers

← Browse all papersPriority research agenda →