Papers › Are NLP Models really able to Solve Simple Math Word Problems?
Are NLP Models really able to Solve Simple Math Word Problems?
Arkil Patel, Satwik Bhattamishra, Navin Goyal
The problem of designing NLP solvers for math word problems (MWP) has seen sustained research activity and steady gains in the test accuracy. Since existing solvers achieve high performance on the benchmark datasets for elementary level MWPs containing one-unknown arithmetic word problems, such problems are often considered "solved" with the bulk of research attention moving to more complex MWPs. In this paper, we restrict our attention to English MWPs taught in grades four and lower. We provide strong evidence that the existing MWP solvers rely on shallow heuristics to achieve high performance on the benchmark datasets. To this end, we show that MWP solvers that do not have access to the question asked in the MWP can still solve a large fraction of MWPs. Similarly, models that treat MWPs as bag-of-words can also achieve surprisingly high accuracy. Further, we introduce a challenge dataset, SVAMP, created by applying carefully chosen variations over examples sampled from existing datasets. The best accuracy achieved by state-of-the-art models is substantially lower on SVAMP, thus showing that much remains to be done even for the simplest of the MWPs.
In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.
Code
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Tasks
Datasets
Introduced by this paper, per the archive.
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Math Word Problem Solving | ASDiv-A | Graph2Tree with RoBERTa | Execution Accuracy | 82.2 | #6 of 9 | Archive leaderboard | report |
| Math Word Problem Solving | ASDiv-A | GTS with RoBERTa | Execution Accuracy | 81.2 | #7 of 9 | Archive leaderboard | report |
| Math Word Problem Solving | ASDiv-A | LSTM Seq2Seq with RoBERTa | Execution Accuracy | 76.9 | #9 of 9 | Archive leaderboard | report |
| Math Word Problem Solving | MAWPS | Graph2Tree with RoBERTa | Accuracy (%) | 88.7 | #10 of 25 | Archive leaderboard | report |
| Math Word Problem Solving | MAWPS | GTS with RoBERTa | Accuracy (%) | 88.5 | #11 of 25 | Archive leaderboard | report |
| Math Word Problem Solving | SVAMP | Graph2Tree with RoBERTa | Accuracy | 43.8 | #21 of 26 | Archive leaderboard | report |
| Math Word Problem Solving | SVAMP | Graph2Tree with RoBERTa | Execution Accuracy | 43.8 | #21 of 26 | Archive leaderboard | report |
| Math Word Problem Solving | SVAMP | GTS with RoBERTa | Accuracy | 41.0 | #22 of 26 | Archive leaderboard | report |
| Math Word Problem Solving | SVAMP | GTS with RoBERTa | Execution Accuracy | 41.0 | #22 of 26 | Archive leaderboard | report |
| Math Word Problem Solving | SVAMP | LSTM Seq2Seq with RoBERTa | Accuracy | 40.3 | #23 of 26 | Archive leaderboard | report |
| Math Word Problem Solving | SVAMP | LSTM Seq2Seq with RoBERTa | Execution Accuracy | 40.3 | #23 of 26 | Archive leaderboard | report |
| Math Word Problem Solving | SVAMP | Transformer with RoBERTa | Accuracy | 38.9 | #25 of 26 | Archive leaderboard | report |
| Math Word Problem Solving | SVAMP | Transformer with RoBERTa | Execution Accuracy | 38.9 | #25 of 26 | Archive leaderboard | report |
Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.
Methods
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections