12,637 papers · updated 10 Oct 2026livingmeta.ai
← Browse all papers
Research theme

AI-Assisted Discovery

The AI-Assisted Discovery theme comprises 1,051 papers in this corpus published between 1982 and 2026. Work here is dominated by Design Science, Empirical Study, Experimental. 5 open research gaps have been surfaced in this area.

Methodology profile

  • Design Science249 (24%)
  • Empirical Study212 (20%)
  • Experimental132 (13%)
  • Case Study106 (10%)
  • Literature Review88 (8%)
  • Benchmarking76 (7%)

Research domains

  • Literature Discovery277 (26%)
  • Data Analysis169 (16%)
  • Research Productivity131 (12%)
  • Human-AI Collaboration127 (12%)
  • Knowledge Synthesis111 (11%)
  • Academic Writing101 (10%)

Frequent sub-topics

LLM-based content coding for competitive dynamics measurement · 1specification format effects when switching from human to AI developers · 1Generative AI for course material navigation and information seeking · 1LLM financial forecasting evaluation · 1AI applications in sustainable public transportation with human-in-the-loop synthesis · 1IS dissertation trends and GenAI-assisted keyword extraction · 1generative AI effects on student learning and metacognition · 1student-AI interaction patterns in learning · 1

Open research gaps

  • not_available
  • Not explicitly stated in the abstract
  • The authors identify that existing LLMs fall short in the financial field due to "disparities between general text data and financial text data," and note that "there is only a limited number of finan
  • The authors identify that "evidence remains limited regarding how rubric design and LLM configurations influence reliability across performance levels." The study also suggests need for further invest
  • The authors identify a gap in understanding preference factors between GenAI tools and traditional search engines for academic help-seeking, and call for improved fluency and reliability of both ChatG

Representative papers

← Browse all papersPriority research agenda →