HyperAI超神経

Mathematical Reasoning On Pgps9K

評価指標

Completion accuracy

評価結果

このベンチマークにおける各モデルのパフォーマンス結果

比較表
モデル名Completion accuracy
a-multi-modal-neural-geometric-solver-with62.7
geoqa-a-geometric-question-answering34.1
inter-gps-interpretable-geometry-problem59.8
unigeo-unifying-geometry-logical-reasoning35.6
gaps-geometry-aware-problem-solver61.2
gold-geometry-problem-solver-with-natural65.8