4 ms·Evaluating frontier AI R&D capabilities of LLM agents against human experts1 points by tedsanders 2y ago