pith. sign in

Generate & rank: A multi-task framework for math word problems

2 Pith papers cite this work. Polarity classification is still indexing.

2 Pith papers citing it

fields

cs.CL 1 cs.LG 1

years

2023 1 2022 1

representative citing papers

Let's Verify Step by Step

cs.LG · 2023-05-31 · accept · novelty 7.0

Process supervision significantly outperforms outcome supervision for training models on the MATH dataset, achieving 78% accuracy on a representative test subset with active learning and a released 800k step-label dataset.

CodeT: Code Generation with Generated Tests

cs.CL · 2022-07-21 · conditional · novelty 7.0

CodeT improves code generation accuracy by using the same model to create test cases and then selecting solutions via output agreement on those tests, raising HumanEval pass@1 from 47% to 65.8%.

citing papers explorer

Showing 2 of 2 citing papers.

  • Let's Verify Step by Step cs.LG · 2023-05-31 · accept · none · ref 16

    Process supervision significantly outperforms outcome supervision for training models on the MATH dataset, achieving 78% accuracy on a representative test subset with active learning and a released 800k step-label dataset.

  • CodeT: Code Generation with Generated Tests cs.CL · 2022-07-21 · conditional · none · ref 11

    CodeT improves code generation accuracy by using the same model to create test cases and then selecting solutions via output agreement on those tests, raising HumanEval pass@1 from 47% to 65.8%.