mixture-of-reasonings-adaptive-strategies

Enable LLMs to autonomously select and apply diverse reasoning strategies without prompt engineering. Trains models with diverse thought templates covering 50-500 distinct reasoning approaches, achieving 2-13% improvements over baseline prompting methods.

Independent GitHub snapshot · Automated classification · Original documentation governs