500AI
Back to researchers

Karl Cobbe

Grade-school math reasoning (GSM8K)

Co-authored GSM8K: a core benchmark/dataset for math word problems and verification.

Highlights

EvaluationMathReasoning
Focus: Grade-school math reasoning (GSM8K)
Why it matters: Co-authored GSM8K: a core benchmark/dataset for math word problems and verification.
Start here
Training Verifiers to Solve Math Word Problems (GSM8K)paper →GSM8K / grade-school-math (dataset)project →

Research Areas

EvaluationMathReasoning
Karl Cobbe - AI Researcher Profile | 500AI