Constrained-CoT
Free while signed in. Answers cite the passages they came from.
First page
The curator’s take
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.
Every Monday
Get next week’s papers.
The same picks and the same summaries, in your inbox. Free, and 176 issues deep.
Subscribe on Substack