By George Tsoukalas and others

We present PutnamBench, a new multilingual benchmark for evaluating the ability of neural theorem-provers to solve competition mathematics problems. PutnamBench consists of 1697 hand-constructed formalizations of 640 theorems sourced from the William Lowell Putnam Mathematical Competition, the premier undergraduate-level mathematics competition in North America. All the theorems have formalizations in... Show more

July 15, 2024

Loading full text...

Similar articles

Loading recommendations...

Artificial IntelligenceComputation and LanguageMachine LearningLogic in Computer ScienceProgramming Languages

x1

PutnamBench: Evaluating Neural Theorem-Provers on the Putnam Mathematical Competition

Click on play to start listening