KAN Author Calls AutoResearch "Monkey Typing," Proposes Meta Models

KAN paper author and MIT PhD student Ziming Liu criticized current LLM-based AutoResearch as "random monkey typing" and proposed Meta Models that can predict training outcomes before training begins, releasing the paper in a follow-up post.

2026-10-31 ~ 2026-10-31 · 2 related posts

• KAN Author Slams AutoResearch as Random Monkey Typing, Proposes Meta Models — ZimingLiu11 · 2026-10-31
• Meta Models Paper Drop: Predicting LLM Training Outcomes Before They Start — ZimingLiu11 · 2026-10-31