ML Research Wiki / Benchmarks / Math Word Problem Solving / MAWPS

MAWPS

Math Word Problem Solving Benchmark

Performance Over Time

📊 Showing 15 results | 📏 Metric: Accuracy (%)

Top Performing Models

Rank Model Paper Accuracy (%) Date Code
1 OpenMath-CodeLlama-70B (w/ code) 📚 OpenMathInstruct-1: A 1.8 Million Math Instruction Tuning Dataset 95.70 2024-02-15 📦 kipok/nemo-skills
2 MsAT-DeductReasoner Learning Multi-Step Reasoning by Solving Arithmetic Tasks 94.30 2023-06-02 📦 TianduoWang/MsAT
3 ATHENA (roberta-large) ATHENA: Mathematical Reasoning with Thought Expansion 93.00 2023-11-02 📦 the-jb/athena-math
4 Multi-view 📚 Multi-View Reasoning: Consistent Contrastive Learning for Math Word Problem 92.30 2022-10-21 📦 zwq2018/multi-view-consistency-for-mwp
5 Exp-Tree An Expression Tree Decoding Strategy for Mathematical Equation Generation 92.30 2023-10-14 📦 zwq2018/multi-view-consistency-for-mwp
6 ATHENA (roberta-base) ATHENA: Mathematical Reasoning with Thought Expansion 92.20 2023-11-02 📦 the-jb/athena-math
7 Roberta-DeductReasoner Learning to Reason Deductively: Math Word Problem Solving as Complex Relation Extraction 92.00 2022-03-19 📦 allanj/deductive-mwp
8 DeBERTa (PM + VM) 📚 Math Word Problem Solving by Generating Linguistic Variants of Problem Statements 91.00 2023-06-24 📦 starscream-11813/variational-mathematical-reasoning
9 Graph2Tree with RoBERTa Are NLP Models really able to Solve Simple Math Word Problems? 88.70 2021-03-12 📦 arkilpatel/SVAMP 📦 debjitpaul/refiner 📦 vedantgaur/symbolic-mwp-reasoning
10 GTS with RoBERTa Are NLP Models really able to Solve Simple Math Word Problems? 88.50 2021-03-12 📦 arkilpatel/SVAMP 📦 debjitpaul/refiner 📦 vedantgaur/symbolic-mwp-reasoning

All Papers (15)