observation · 17 Oct 2025 · 50:00

Current research papers on synthetic examples for training LLMs are often just ideas and lack convincing demonstrations at frontier LLM scale.

I am only at a stage of reading abstracts because a lot of these papers are just ideas. Someone has to make it work on a frontier LLM lab scale in full generality because when you see these papers, they pop up, and it's just a bit noisy. They're cool ideas, but I haven't seen anyone convincingly show that this is possible.

Watch at 50:00