Optimization researcher: I couldn't have derived these new constructions without LLMs
ErnestRyu · x · 2026-08-19
Optimization researcher Ernest Ryu describes the new algorithms found by the LLM-driven AutoOPT — the Lemniscate Accelerated Gradient Method and Analytic ITEM-f. He admits the constructions are "frightening": without LLMs and the domain-specific harness, he would not have figured them out on his own.
Related event: LLM-Powered AutoOPT Discovers New Optimization Algorithms(3 posts)→
More from Research
- Stop treating Platonic representation hypothesis as fact — jiaxinwen22 · 2026-08-19
- Study: LLM Judges Are Useless Below Moderate Agreement — IanArawjo · 2026-08-19
- LlamaParse Outperforms General VLMs in Document Visual Grounding — llama_index · 2026-08-19
- Professor tests AIs on an open conjecture — all agree a general proof remains elusive — lemire · 2026-08-19
- GenBio releases AIDO Cell, the first multi-modal world model of a human cell — HanGuo97 · 2026-08-19
- Paper Proposes GMAD Loss to Replace RMSE for ML Trading Model Optimization — PtrPomorski · 2026-08-19