Georg Unterholzner @georg.dev · Jun 11 Large Language Models (LLMs) with "reasoning" capabilities are super hyped, but new research from Apple shows their "thinking" is often an illusion. Beyond a certain complexity, they fail completely. Key points in the 🧵.
2 likes 1 replies
?
Replies
Key points: - At low complexity, standard LLMs excel (faster, cheaper). - At medium complexity, reasoning models (LRMs) outperform. - At high complexity, both collapse, with nearly 0% accuracy.