Hybrid model w/ SNI

Reported on 4 benchmarks across 4 tasks

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

Knowledge Base2 results

Mathematical Question AnsweringonMath23K
Accuracy (5-fold)
64.7
best: 94.3 (GPT-4 (Teaching-Inspired))
Mathematical ReasoningonMath23K
Accuracy (5-fold)
64.7
best: 94.3 (GPT-4 (Teaching-Inspired))

Natural Language Processing1 result

Question AnsweringonMath23K
Accuracy (5-fold)
64.7
best: 94.3 (GPT-4 (Teaching-Inspired))

Reasoning1 result

Math Word Problem SolvingonMath23K
Accuracy (5-fold)
64.7
best: 94.3 (GPT-4 (Teaching-Inspired))