# Math Word Problem Generation on Standards-aligned Educational MWPs

Source: [Math Word Problem Generation on Standards-aligned Educational MWPs](<https://www.sota2.com/research/sota/math-word-problem-generation-on-standards-aligned-educational-mwps>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-04-14T08:32:32.582Z

Evaluation of model performance on generating math word problems using the Standards-aligned Educational MWPs dataset, reporting generation quality (PPL), linguistic features (Readability, Length), and specific evaluation scores (BF1, MaC).

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Math Word Problem Generation |
| Dataset (including any stated split) | Standards-aligned Educational MWPs |
| Metric | PPL |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 9 of 9 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [PPL](<https://www.sota2.com/research/sota/math-word-problem-generation-on-standards-aligned-educational-mwps.md?metric=PPL>): Lower is better; unit: SCALAR.

- [BF1](<https://www.sota2.com/research/sota/math-word-problem-generation-on-standards-aligned-educational-mwps.md?metric=BF1>): Higher is better; unit: PERCENT.

- [ASDIV BF1](<https://www.sota2.com/research/sota/math-word-problem-generation-on-standards-aligned-educational-mwps.md?metric=ASDIV%20BF1>): Higher is better; unit: PERCENT.

- [Question Length](<https://www.sota2.com/research/sota/math-word-problem-generation-on-standards-aligned-educational-mwps.md?metric=Question%20Length>): Higher is better; unit: SCALAR.

- [Solution Length](<https://www.sota2.com/research/sota/math-word-problem-generation-on-standards-aligned-educational-mwps.md?metric=Solution%20Length>): Higher is better; unit: SCALAR.

- [Question Readability](<https://www.sota2.com/research/sota/math-word-problem-generation-on-standards-aligned-educational-mwps.md?metric=Question%20Readability>): Higher is better; unit: SCALAR.

- [Solution Readability](<https://www.sota2.com/research/sota/math-word-problem-generation-on-standards-aligned-educational-mwps.md?metric=Solution%20Readability>): Higher is better; unit: SCALAR.

- [MaC](<https://www.sota2.com/research/sota/math-word-problem-generation-on-standards-aligned-educational-mwps.md?metric=MaC>): Higher is better; unit: PERCENT.

## Recorded results

| Method | PPL | Result date | Report paper |
| --- | --- | --- | --- |
| EDUMATH 12B | 9.5 | Not recorded | 68fa7bed1e209131cea8b5af |
| Gemma 3 12B IT | 11.3 | Not recorded | 68fa7bed1e209131cea8b5af |
| EDUMATH 30B | 12 | Not recorded | 68fa7bed1e209131cea8b5af |
| Gemma 3 27B IT | 12.2 | Not recorded | 68fa7bed1e209131cea8b5af |
| Qwen 3 30B IT | 12.3 | Not recorded | 68fa7bed1e209131cea8b5af |
| Qwen 3 235B IT | 12.5 | Not recorded | 68fa7bed1e209131cea8b5af |
| GPT-4o-mini | 15.7 | Not recorded | 68fa7bed1e209131cea8b5af |
| GPT-4o | 16.1 | Not recorded | 68fa7bed1e209131cea8b5af |
| GPT-4.0 | 16.3 | Not recorded | 68fa7bed1e209131cea8b5af |

### EDUMATH 12B

PPL: 9.5. Result ID: 69ddfb9a47f34ce924c56abd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b5af.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b5af.md>)

Recorded evaluation configuration: {"Access Type":"Public","Model Size":"12B"}. This record does not assert independent reproduction.

### Gemma 3 12B IT

PPL: 11.3. Result ID: 69ddfb9a47f34ce924c56ab9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b5af.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa779f1e209131cea746cd.md>)

Recorded evaluation configuration: {"Access Type":"Public","Model Size":"12B","Instruction Tuned":"true"}. This record does not assert independent reproduction.

### EDUMATH 30B

PPL: 12. Result ID: 69ddfb9a47f34ce924c56abe.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b5af.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b5af.md>)

Recorded evaluation configuration: {"Access Type":"Public","Model Size":"30B"}. This record does not assert independent reproduction.

### Gemma 3 27B IT

PPL: 12.2. Result ID: 69ddfb9a47f34ce924c56aba.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b5af.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75001e209131cea6527c.md>)

Recorded evaluation configuration: {"Access Type":"Public","Model Size":"27B","Instruction Tuned":"true"}. This record does not assert independent reproduction.

### Qwen 3 30B IT

PPL: 12.3. Result ID: 69ddfb9a47f34ce924c56abb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b5af.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Access Type":"Public","Model Size":"30B","Instruction Tuned":"true"}. This record does not assert independent reproduction.

### Qwen 3 235B IT

PPL: 12.5. Result ID: 69ddfb9a47f34ce924c56abc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b5af.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Access Type":"Public","Model Size":"235B","Instruction Tuned":"true"}. This record does not assert independent reproduction.

### GPT-4o-mini

PPL: 15.7. Result ID: 69ddfb9a47f34ce924c56ab8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b5af.md>)

Recorded evaluation configuration: {"Access Type":"API"}. This record does not assert independent reproduction.

### GPT-4o

PPL: 16.1. Result ID: 69ddfb9a47f34ce924c56ab6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b5af.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75681e209131cea67a98.md>)

Recorded evaluation configuration: {"Access Type":"API"}. This record does not assert independent reproduction.

### GPT-4.0

PPL: 16.3. Result ID: 69ddfb9a47f34ce924c56ab7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7bed1e209131cea8b5af.md>)

Recorded evaluation configuration: {"Access Type":"API"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
