# Paragraph Segmentation on TEDPara (val)

Source: [Paragraph Segmentation on TEDPara (val)](<https://www.sota2.com/research/sota/paragraph-segmentation-on-tedpara-val>)

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

Data updated: 2026-02-26T05:43:32.940Z

Evaluation of paragraph segmentation performance on the validation split of the TEDPara dataset, reporting F1, BS, and Pk 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 | Paragraph Segmentation |
| Dataset (including any stated split) | TEDPara (val) |
| Metric | F1 Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 16 of 16 |
| 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 Score](<https://www.sota2.com/research/sota/paragraph-segmentation-on-tedpara-val.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

- [Boundary Score (BS)](<https://www.sota2.com/research/sota/paragraph-segmentation-on-tedpara-val.md?metric=Boundary%20Score%20(BS)>): Higher is better; unit: PERCENT.

- [Pk Score](<https://www.sota2.com/research/sota/paragraph-segmentation-on-tedpara-val.md?metric=Pk%20Score>): Lower is better; unit: PERCENT.

## Recorded results

| Method | F1 Score | Result date | Report paper |
| --- | --- | --- | --- |
| MiniSeg | 72.1 | Not recorded | 6955ff8e53fdd5c36173e012 |
| MiniSeg | 70.6 | Not recorded | 6955ff8e53fdd5c36173e012 |
| MiniSeg | 69.8 | Not recorded | 6955ff8e53fdd5c36173e012 |
| MiniSeg | 67.3 | Not recorded | 6955ff8e53fdd5c36173e012 |
| LLaMA 3.1 70B | 55.5 | Not recorded | 6955ff8e53fdd5c36173e012 |
| LLaMA 3.3 70B | 52.2 | Not recorded | 6955ff8e53fdd5c36173e012 |
| LLaMA 3.1 8B | 51.7 | Not recorded | 6955ff8e53fdd5c36173e012 |
| MiniSeg | 45.7 | Not recorded | 6955ff8e53fdd5c36173e012 |
| LLaMA 3.1 70B | 42.8 | Not recorded | 6955ff8e53fdd5c36173e012 |
| LLaMA 3.3 70B | 37.5 | Not recorded | 6955ff8e53fdd5c36173e012 |
| LLaMA 3.1 8B | 36.6 | Not recorded | 6955ff8e53fdd5c36173e012 |
| MiniSeg | 33.2 | Not recorded | 6955ff8e53fdd5c36173e012 |
| MiniSeg | 30.5 | Not recorded | 6955ff8e53fdd5c36173e012 |
| MiniSeg | 24.6 | Not recorded | 6955ff8e53fdd5c36173e012 |
| Rule-Based Baseline | 21.6 | Not recorded | 6955ff8e53fdd5c36173e012 |
| Random Baseline | 16.8 | Not recorded | 6955ff8e53fdd5c36173e012 |

### MiniSeg

F1 Score: 72.1. Result ID: 6996a9675698665ec4f34ae6.

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

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

Recorded evaluation configuration: {"Evaluation Protocol":"Fine-Tuned","Training Pipeline":"Wiki → YT → TED"}. This record does not assert independent reproduction.

### MiniSeg

F1 Score: 70.6. Result ID: 6996a9675698665ec4f34ae5.

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

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

Recorded evaluation configuration: {"Evaluation Protocol":"Fine-Tuned","Training Pipeline":"Wiki → TED"}. This record does not assert independent reproduction.

### MiniSeg

F1 Score: 69.8. Result ID: 6996a9675698665ec4f34ae4.

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

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

Recorded evaluation configuration: {"Evaluation Protocol":"Fine-Tuned","Training Pipeline":"YT → TED"}. This record does not assert independent reproduction.

### MiniSeg

F1 Score: 67.3. Result ID: 6996a9675698665ec4f34ae3.

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

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

Recorded evaluation configuration: {"Evaluation Protocol":"Fine-Tuned","Training Data":"TEDPara"}. This record does not assert independent reproduction.

### LLaMA 3.1 70B

F1 Score: 55.5. Result ID: 6996a9675698665ec4f34ae1.

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

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

Recorded evaluation configuration: {"Evaluation Protocol":"Zero-Shot","Paralinguistic Break Rule (PBR)":true}. This record does not assert independent reproduction.

### LLaMA 3.3 70B

F1 Score: 52.2. Result ID: 6996a9675698665ec4f34ae2.

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

[Method origin paper](<https://www.sota2.com/research/paper/68fa75d11e209131cea6a028.md>)

[Implementation](<https://github.com/meta-llama/llama-agentic-system>)

Recorded evaluation configuration: {"Evaluation Protocol":"Zero-Shot","Paralinguistic Break Rule (PBR)":true}. This record does not assert independent reproduction.

### LLaMA 3.1 8B

F1 Score: 51.7. Result ID: 6996a9675698665ec4f34ae0.

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

[Method origin paper](<https://www.sota2.com/research/paper/68fa75d11e209131cea6a028.md>)

[Implementation](<https://github.com/meta-llama/llama-agentic-system>)

Recorded evaluation configuration: {"Evaluation Protocol":"Zero-Shot","Paralinguistic Break Rule (PBR)":true}. This record does not assert independent reproduction.

### MiniSeg

F1 Score: 45.7. Result ID: 6996a9675698665ec4f34adf.

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

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

Recorded evaluation configuration: {"Evaluation Protocol":"Zero-Shot","Pre-training":"Wiki → YT","Threshold Tuning":true}. This record does not assert independent reproduction.

### LLaMA 3.1 70B

F1 Score: 42.8. Result ID: 6996a9675698665ec4f34ada.

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

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

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

### LLaMA 3.3 70B

F1 Score: 37.5. Result ID: 6996a9675698665ec4f34adb.

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

[Method origin paper](<https://www.sota2.com/research/paper/68fa75d11e209131cea6a028.md>)

[Implementation](<https://github.com/meta-llama/llama-agentic-system>)

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

### LLaMA 3.1 8B

F1 Score: 36.6. Result ID: 6996a9675698665ec4f34ad9.

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

[Method origin paper](<https://www.sota2.com/research/paper/68fa75d11e209131cea6a028.md>)

[Implementation](<https://github.com/meta-llama/llama-agentic-system>)

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

### MiniSeg

F1 Score: 33.2. Result ID: 6996a9675698665ec4f34ade.

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

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

Recorded evaluation configuration: {"Evaluation Protocol":"Zero-Shot","Pre-training":"Wiki → YT"}. This record does not assert independent reproduction.

### MiniSeg

F1 Score: 30.5. Result ID: 6996a9675698665ec4f34add.

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

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

Recorded evaluation configuration: {"Evaluation Protocol":"Zero-Shot","Pre-training":"Wiki","Threshold Tuning":true}. This record does not assert independent reproduction.

### MiniSeg

F1 Score: 24.6. Result ID: 6996a9675698665ec4f34adc.

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

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

Recorded evaluation configuration: {"Evaluation Protocol":"Zero-Shot","Pre-training":"Wiki"}. This record does not assert independent reproduction.

### Rule-Based Baseline

F1 Score: 21.6. Result ID: 6996a9675698665ec4f34ad8.

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

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

### Random Baseline

F1 Score: 16.8. Result ID: 6996a9675698665ec4f34ad7.

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