Can AI Evaluate AI Scientists? A Benchmarking Study of Autonomous Research Generation Systems Using Automated⭐9

Benchmarking AI Scientist systems via automated multi-model review reveals FARS papers score 2x higher across originality, rigor, clarity, and significance.