Solving math word problems
OpenAI Blog
A research team developed a system that solves grade school math word problems at higher accuracy than fine-tuned GPT-3. The system achieved 55% accuracy on problems where a small sample of 9-12 year olds scored 60%. The improvement demonstrates progress in training models to handle mathematical reasoning tasks that require multi-step problem solving.
Why it matters
We’ve trained a system that solves grade school math problems with nearly twice the accuracy of a fine-tuned GPT-3 model. It solves about 90% as many problems as real kids: a small sample of 9-12 year olds scored 60% on a test from our dataset, while our system scored 55% on those same problems.