(논문 요약) Towards Robust Mathematical Reasoning (Paper)

핵심 내용

• IMO-Bench
  • IMO-AnswerBench: 400 diverse Olympiad problems with verifiable short answers.
  • IMO-Proof Bench: 60 problems to evaluate proof-writing capabilities.
  • IMO-GradingBench: 1000 human gradings on proofs.
    • ProofAutoGrader: leverages Gemini 2.5 Pro, providing it with a prompt containing the problem statement, the candidate solution, a reference solution, and specific grading guidelines