Language Models vs. Human Experts on Professional Knowledge and Science Benchmarks

This report synthesises findings from 12 peer-reviewed papers addressing the following research question: How do language models compare to human experts on professional knowledge and science benchmarks v8. 9 claims were extracted from source literature; 9 were independently verified against retrieved documents. An automated multi-reviewer quality assessment produced a score of 9.2/10. This report is a machine-generated literature synthesis and does not constitute original research.

Research goal: How do language models compare to human experts on professional knowledge and science benchmarks v8

Autonomous literature synthesis. Automated review score: 9.2/10. Full text and citation available at Assignee Research.