Minerva, a language model trained on technical content, achieves state-of-the-art performance on quantitative reasoning benchmarks and solves nearly a third of undergraduate-level science and math problems.
Model output (62B):The sum of the geometric series is 0.12 1−0.12 = 1/2 1−1/2 = 1/2 1/2 = 1
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2022 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Solving Quantitative Reasoning Problems with Language Models
Minerva, a language model trained on technical content, achieves state-of-the-art performance on quantitative reasoning benchmarks and solves nearly a third of undergraduate-level science and math problems.