# Punctuation Restoration on PersianPunc (test)

Source: [Punctuation Restoration on PersianPunc (test)](<https://www.sota2.com/research/sota/punctuation-restoration-on-persianpunc-test>)

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

Data updated: 2026-03-06T09:43:47.591Z

Evaluation of model performance on the test split of the PersianPunc dataset for the task of Punctuation Restoration, reporting Macro F1 and FSM scores.

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 | Punctuation Restoration |
| Dataset (including any stated split) | PersianPunc (test) |
| Metric | Macro F1 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 3 of 3 |
| 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

- [Macro F1](<https://www.sota2.com/research/sota/punctuation-restoration-on-persianpunc-test.md?metric=Macro%20F1>): Higher is better; unit: PERCENT.

- [FSM](<https://www.sota2.com/research/sota/punctuation-restoration-on-persianpunc-test.md?metric=FSM>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Macro F1 | Result date | Report paper |
| --- | --- | --- | --- |
| ParsBERT | 91.33 | Not recorded | 69aa5f962d603e6de3e419dd |
| GPT-4o | 85.96 | Not recorded | 69aa5f962d603e6de3e419dd |
| GPT-4o-mini | 79.54 | Not recorded | 69aa5f962d603e6de3e419dd |

### ParsBERT

Macro F1: 91.33. Result ID: 69aaa1c96dbb7075b954c138.

[Reporting paper](<https://www.sota2.com/research/paper/69aa5f962d603e6de3e419dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69aa5f962d603e6de3e419dd.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"fine-tuned"}. This record does not assert independent reproduction.

### GPT-4o

Macro F1: 85.96. Result ID: 69aaa1c96dbb7075b954c137.

[Reporting paper](<https://www.sota2.com/research/paper/69aa5f962d603e6de3e419dd.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa727a1e209131cea56a1e.md>)

[Implementation](<https://github.com/openai/evals>)

Recorded evaluation configuration: {"Evaluation Protocol":"zero-shot"}. This record does not assert independent reproduction.

### GPT-4o-mini

Macro F1: 79.54. Result ID: 69aaa1c96dbb7075b954c136.

[Reporting paper](<https://www.sota2.com/research/paper/69aa5f962d603e6de3e419dd.md>)

Recorded evaluation configuration: {"Evaluation Protocol":"zero-shot"}. This record does not assert independent reproduction.

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