Core Concepts
Enhancing reasoning capabilities of large language models through improved CoT prompting.
Stats
"Extensive experiments demonstrate that our ChainLM models exhibit enhanced proficiency in addressing a spectrum of complex reasoning problems compared to existing models."
"We repeat the improvement process for four rounds using OpenAI ChatGPT API and finally obtain 44,335 CoT prompts."
Quotes
"Chain-of-Thought (CoT) prompting has been proposed and emerged as an effective solution for complex reasoning."
"Our ChainLM model exhibited better performance when confronted with complex reasoning tasks."