GrapeQA: PEGA

Reported on 3 benchmarks across 2 tasks · 1 paper

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

Natural Language Processing3 results

Question AnsweringonMedQA
Accuracy· 2023-03-22
39.51
best: 91.1 (Med-Gemini)
GrapeQA: GRaph Augmentation and Pruning to Enhance Question-Answering arXiv:2303.12320
Question AnsweringonOpenBookQA
Accuracy· 2023-03-22
82
best: 95.9 (GPT-4 + knowledge base)
GrapeQA: GRaph Augmentation and Pruning to Enhance Question-Answering arXiv:2303.12320
Common Sense ReasoningonCommonsenseQA
Accuracy· 2023-03-22
73.5
best: 92.54 (GPT-4o (HPT))
GrapeQA: GRaph Augmentation and Pruning to Enhance Question-Answering arXiv:2303.12320