# Paragraph Segmentation on YTSegPara (test)

Source: [Paragraph Segmentation on YTSegPara (test)](<https://www.sota2.com/research/sota/paragraph-segmentation-on-ytsegpara-test>)

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

Data updated: 2026-02-26T05:42:46.918Z

Evaluation of Paragraph Segmentation performance on the test split of the YTSegPara dataset, reporting F1, BS, and Pk metrics.

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 | Paragraph Segmentation |
| Dataset (including any stated split) | YTSegPara (test) |
| Metric | F1 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [F1](<https://www.sota2.com/research/sota/paragraph-segmentation-on-ytsegpara-test.md?metric=F1>): Higher is better; unit: PERCENT.

- [BS](<https://www.sota2.com/research/sota/paragraph-segmentation-on-ytsegpara-test.md?metric=BS>): Higher is better; unit: PERCENT.

- [Pk](<https://www.sota2.com/research/sota/paragraph-segmentation-on-ytsegpara-test.md?metric=Pk>): Higher is better; unit: PERCENT.

## Recorded results

| Method | F1 | Result date | Report paper |
| --- | --- | --- | --- |
| MiniSeg (Wiki → YT) | 50.5 | Not recorded | 6955ff8e53fdd5c36173e012 |
| MiniSeg (YT) | 47.9 | Not recorded | 6955ff8e53fdd5c36173e012 |
| MiniSeg (Wiki → TED) | 35.7 | Not recorded | 6955ff8e53fdd5c36173e012 |
| Random Baseline | 26.3 | Not recorded | 6955ff8e53fdd5c36173e012 |

### MiniSeg (Wiki → YT)

F1: 50.5. Result ID: 6996a9675698665ec4f34ad1.

[Reporting paper](<https://www.sota2.com/research/paper/6955ff8e53fdd5c36173e012.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6955ff8e53fdd5c36173e012.md>)

Recorded evaluation configuration: {"Training Strategy":"Hierarchical"}. This record does not assert independent reproduction.

### MiniSeg (YT)

F1: 47.9. Result ID: 6996a9675698665ec4f34ad0.

[Reporting paper](<https://www.sota2.com/research/paper/6955ff8e53fdd5c36173e012.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6955ff8e53fdd5c36173e012.md>)

Recorded evaluation configuration: {"Training Strategy":"Hierarchical"}. This record does not assert independent reproduction.

### MiniSeg (Wiki → TED)

F1: 35.7. Result ID: 6996a9675698665ec4f34acf.

[Reporting paper](<https://www.sota2.com/research/paper/6955ff8e53fdd5c36173e012.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6955ff8e53fdd5c36173e012.md>)

Recorded evaluation configuration: {"Training Strategy":"Single-task","Evaluation Context":"Scored within oracle chapter spans"}. This record does not assert independent reproduction.

### Random Baseline

F1: 26.3. Result ID: 6996a9675698665ec4f34ace.

[Reporting paper](<https://www.sota2.com/research/paper/6955ff8e53fdd5c36173e012.md>)

Recorded evaluation configuration: {"Note":"oracle knowledge of the number of chapters and paragraph-break rate"}. This record does not assert independent reproduction.

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