# Structured tax-form completion on TaxCalcBench TY24 (test)

Source: [Structured tax-form completion on TaxCalcBench TY24 (test)](<https://www.sota2.com/research/sota/structured-tax-form-completion-on-taxcalcbench-ty24-test>)

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

Data updated: 2026-05-12T08:00:27.514Z

Evaluation of model performance on structured tax-form completion using the test split of the TaxCalcBench TY24 dataset, measured across specific case accuracies.

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 | Structured tax-form completion |
| Dataset (including any stated split) | TaxCalcBench TY24 (test) |
| Metric | Case 1 Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 10 of 10 |
| 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

- [Case 1 Accuracy](<https://www.sota2.com/research/sota/structured-tax-form-completion-on-taxcalcbench-ty24-test.md?metric=Case%201%20Accuracy>): Higher is better; unit: PERCENT.

- [Case 3 Accuracy](<https://www.sota2.com/research/sota/structured-tax-form-completion-on-taxcalcbench-ty24-test.md?metric=Case%203%20Accuracy>): Higher is better; unit: PERCENT.

- [Case 4 Accuracy](<https://www.sota2.com/research/sota/structured-tax-form-completion-on-taxcalcbench-ty24-test.md?metric=Case%204%20Accuracy>): Higher is better; unit: PERCENT.

- [Case 6a Accuracy](<https://www.sota2.com/research/sota/structured-tax-form-completion-on-taxcalcbench-ty24-test.md?metric=Case%206a%20Accuracy>): Higher is better; unit: PERCENT.

- [Test Accuracy](<https://www.sota2.com/research/sota/structured-tax-form-completion-on-taxcalcbench-ty24-test.md?metric=Test%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Case 1 Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| DAPO | 55 | Not recorded | 6a02b3f8dbc836b87d82459d |
| GRPO | 54 | Not recorded | 6a02b3f8dbc836b87d82459d |
| Zero-shot | 53 | Not recorded | 6a02b3f8dbc836b87d82459d |
| GRPO | 52 | Not recorded | 6a02b3f8dbc836b87d82459d |
| GSPO | 52 | Not recorded | 6a02b3f8dbc836b87d82459d |
| DPA-GRPO | 52 | Not recorded | 6a02b3f8dbc836b87d82459d |
| DAPO | 47 | Not recorded | 6a02b3f8dbc836b87d82459d |
| GSPO | 46 | Not recorded | 6a02b3f8dbc836b87d82459d |
| DPA-GRPO | 43 | Not recorded | 6a02b3f8dbc836b87d82459d |
| Zero-shot | 41 | Not recorded | 6a02b3f8dbc836b87d82459d |

### DAPO

Case 1 Accuracy: 55. Result ID: 6a02dce347f34ce924e56e04.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b3f8dbc836b87d82459d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78a81e209131cea79ac6.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-8B"}. This record does not assert independent reproduction.

### GRPO

Case 1 Accuracy: 54. Result ID: 6a02dce347f34ce924e56e03.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b3f8dbc836b87d82459d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73101e209131cea5a395.md>)

[Implementation](<https://github.com/deepseek-ai/DeepSeek-Math>)

Recorded evaluation configuration: {"Backbone":"Qwen3-8B"}. This record does not assert independent reproduction.

### Zero-shot

Case 1 Accuracy: 53. Result ID: 6a02dce347f34ce924e56dff.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b3f8dbc836b87d82459d.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-8B"}. This record does not assert independent reproduction.

### GRPO

Case 1 Accuracy: 52. Result ID: 6a02dce347f34ce924e56e00.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b3f8dbc836b87d82459d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73101e209131cea5a395.md>)

[Implementation](<https://github.com/deepseek-ai/DeepSeek-Math>)

Recorded evaluation configuration: {"Backbone":"Qwen3-4B"}. This record does not assert independent reproduction.

### GSPO

Case 1 Accuracy: 52. Result ID: 6a02dce347f34ce924e56e05.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b3f8dbc836b87d82459d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a481e209131cea8231d.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-8B"}. This record does not assert independent reproduction.

### DPA-GRPO

Case 1 Accuracy: 52. Result ID: 6a02dce347f34ce924e56e06.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b3f8dbc836b87d82459d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b3f8dbc836b87d82459d.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-4B"}. This record does not assert independent reproduction.

### DAPO

Case 1 Accuracy: 47. Result ID: 6a02dce347f34ce924e56e01.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b3f8dbc836b87d82459d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78a81e209131cea79ac6.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-4B"}. This record does not assert independent reproduction.

### GSPO

Case 1 Accuracy: 46. Result ID: 6a02dce347f34ce924e56e02.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b3f8dbc836b87d82459d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a481e209131cea8231d.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-4B"}. This record does not assert independent reproduction.

### DPA-GRPO

Case 1 Accuracy: 43. Result ID: 6a02dce347f34ce924e56e07.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b3f8dbc836b87d82459d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b3f8dbc836b87d82459d.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-8B"}. This record does not assert independent reproduction.

### Zero-shot

Case 1 Accuracy: 41. Result ID: 6a02dce347f34ce924e56dfe.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b3f8dbc836b87d82459d.md>)

Recorded evaluation configuration: {"Backbone":"Qwen3-4B"}. This record does not assert independent reproduction.

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