4/ Constrained-CoT – limits the model reasoning output length without sacrificing performance; shows that constraining the reasoning of LLaMA2-70b to 100 words improves the accuracy from 36.01% (CoT) to 41.07% (CCoT) on GSM8K, while reducing the average output length by 28 words.
Constrained Chain-of-Thought Improves LLaMA2 Reasoning Accuracy
By
–