TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/Shepherd + DeepSeek-67B (SFT on MetaMATH + PRM rerank, k=256)

Shepherd + DeepSeek-67B (SFT on MetaMATH + PRM rerank, k=256)

Reported on 8 benchmarks across 4 tasks · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Knowledge Base4 results

  • Mathematical Question AnsweringonMATH
    Accuracy· uses extra data· 2023-12-14
    48.1
    best: 89.7 (Gemini 2.0 Flash Experimental)
    Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human AnnotationsarXiv:2312.08935
  • Mathematical Question AnsweringonMATH
    Parameters (Billions)· uses extra data· 2023-12-14
    67
    best: 540 (Minerva 540B)
    Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human AnnotationsarXiv:2312.08935
  • Mathematical ReasoningonMATH
    Accuracy· uses extra data· 2023-12-14
    48.1
    best: 89.7 (Gemini 2.0 Flash Experimental)
    Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human AnnotationsarXiv:2312.08935
  • Mathematical ReasoningonMATH
    Parameters (Billions)· uses extra data· 2023-12-14
    67
    best: 540 (Minerva 540B)
    Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human AnnotationsarXiv:2312.08935

Natural Language Processing2 results

  • Question AnsweringonMATH
    Accuracy· uses extra data· 2023-12-14
    48.1
    best: 89.7 (Gemini 2.0 Flash Experimental)
    Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human AnnotationsarXiv:2312.08935
  • Question AnsweringonMATH
    Parameters (Billions)· uses extra data· 2023-12-14
    67
    best: 540 (Minerva 540B)
    Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human AnnotationsarXiv:2312.08935

Reasoning2 results

  • Math Word Problem SolvingonMATH
    Accuracy· uses extra data· 2023-12-14
    48.1
    best: 89.7 (Gemini 2.0 Flash Experimental)
    Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human AnnotationsarXiv:2312.08935
  • Math Word Problem SolvingonMATH
    Parameters (Billions)· uses extra data· 2023-12-14
    67
    best: 540 (Minerva 540B)
    Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human AnnotationsarXiv:2312.08935