search

hkust-nlp / dart-math

[NeurIPS'24] Official code for *🎯DART-Math: Difficulty-Aware Rejection Tuning for Mathematical Problem-Solving*
https://hkust-nlp.github.io/dart-math/
MIT License
65 stars 3 forks source link

nice work. could you provide amc2023 and aime 2024 dataset and evaluation script #1

Closed yyht closed 2 months ago

yyht commented 3 months ago

nice work. will you release amc2023 and aime 2024 dataset and evaluation script which was posted on https://x.com/tongyx361/status/1815112376649134172

tongyx361 commented 2 months ago

Sorry that the evaluation on AMC2023 and AIME2024 is conducted by the Numina team, so I don't have the exact evaluation datasets and scripts.

You can try

  1. opening an issue in their repo.,
  2. constructing your own versions with the AOPS dataset provided in the repo., which does not include the problems in 2024,
  3. crawling from the AOPS website.