# Multi-class classification on PROMISE original (10-fold cross-validation)

Source: [Multi-class classification on PROMISE original (10-fold cross-validation)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation>)

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

Data updated: 2026-06-04T06:49:22.041Z

Evaluation of Multi-class Classification performance on the PROMISE original dataset using a 10-fold cross-validation scheme, reporting per-class Precision, Recall, and F1 scores for various categories (FR, A, FT, etc.), along with overall averages.

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 | Multi-class classification |
| Dataset (including any stated split) | PROMISE original (10-fold cross-validation) |
| Metric | F1 Score (A) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 13 of 13 |
| 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 (A)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=F1%20Score%20(A)>): Higher is better; unit: PERCENT.

- [F1 Score (FR)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=F1%20Score%20(FR)>): Higher is better; unit: PERCENT.

- [Precision (FR)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=Precision%20(FR)>): Higher is better; unit: PERCENT.

- [Recall (FR)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=Recall%20(FR)>): Higher is better; unit: PERCENT.

- [Precision (A)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=Precision%20(A)>): Higher is better; unit: PERCENT.

- [Recall (A)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=Recall%20(A)>): Higher is better; unit: PERCENT.

- [F1 Score (FT)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=F1%20Score%20(FT)>): Higher is better; unit: PERCENT.

- [Precision (FT)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=Precision%20(FT)>): Higher is better; unit: PERCENT.

- [Recall (FT)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=Recall%20(FT)>): Higher is better; unit: PERCENT.

- [F1 Score (L)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=F1%20Score%20(L)>): Higher is better; unit: PERCENT.

- [Precision (L)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=Precision%20(L)>): Higher is better; unit: PERCENT.

- [Recall (L)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=Recall%20(L)>): Higher is better; unit: PERCENT.

- [F1 Score (LF)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=F1%20Score%20(LF)>): Higher is better; unit: PERCENT.

- [Precision (LF)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=Precision%20(LF)>): Higher is better; unit: PERCENT.

- [Recall (LF)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=Recall%20(LF)>): Higher is better; unit: PERCENT.

- [F1 Score (MN)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=F1%20Score%20(MN)>): Higher is better; unit: PERCENT.

- [Precision (MN)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=Precision%20(MN)>): Higher is better; unit: PERCENT.

- [Recall (MN)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=Recall%20(MN)>): Higher is better; unit: PERCENT.

- [F1 Score (O)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=F1%20Score%20(O)>): Higher is better; unit: PERCENT.

- [Precision (O)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=Precision%20(O)>): Higher is better; unit: PERCENT.

- [Recall (O)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=Recall%20(O)>): Higher is better; unit: PERCENT.

- [F1 Score (PE)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=F1%20Score%20(PE)>): Higher is better; unit: PERCENT.

- [Precision (PE)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=Precision%20(PE)>): Higher is better; unit: PERCENT.

- [Recall (PE)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=Recall%20(PE)>): Higher is better; unit: SCALAR.

- [F1 Score (SC)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=F1%20Score%20(SC)>): Higher is better; unit: PERCENT.

- [Precision (SC)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=Precision%20(SC)>): Higher is better; unit: PERCENT.

- [Recall (SC)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=Recall%20(SC)>): Higher is better; unit: PERCENT.

- [F1 Score (SE)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=F1%20Score%20(SE)>): Higher is better; unit: PERCENT.

- [Precision (SE)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=Precision%20(SE)>): Higher is better; unit: PERCENT.

- [Recall (SE)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=Recall%20(SE)>): Higher is better; unit: SCALAR.

- [F1 Score (US)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=F1%20Score%20(US)>): Higher is better; unit: PERCENT.

- [Precision (US)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=Precision%20(US)>): Higher is better; unit: PERCENT.

- [Recall (US)](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=Recall%20(US)>): Higher is better; unit: PERCENT.

- [Average Score](<https://www.sota2.com/research/sota/multi-class-classification-on-promise-original-10-fold-cross-validation.md?metric=Average%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | F1 Score (A) | Result date | Report paper |
| --- | --- | --- | --- |
| DeepSeek-V3 | 100 | Not recorded | 68fa7b661e209131cea88313 |
| PRC-RoBERTa-L | 100 | Not recorded | 68fa7b661e209131cea88313 |
| GPT-3.5-turbo | 94 | Not recorded | 68fa7b661e209131cea88313 |
| Mistral-7B-ins. | 93 | Not recorded | 68fa7b661e209131cea88313 |
| PRC-BERT-B | 93 | Not recorded | 68fa7b661e209131cea88313 |
| NB (DP) | 90 | Not recorded | 68fa7b661e209131cea88313 |
| LLaMA-3.1-8B-ins. | 81 | Not recorded | 68fa7b661e209131cea88313 |
| NoR-BERT-B | 79 | Not recorded | 68fa7b661e209131cea88313 |
| NoR-BERT-L | 78 | Not recorded | 68fa7b661e209131cea88313 |
| LDA (DP) | 74 | Not recorded | 68fa7b661e209131cea88313 |
| GPT-4o | 68 | Not recorded | 68fa7b661e209131cea88313 |
| Gemma-3-4B | 67 | Not recorded | 68fa7b661e209131cea88313 |
| LLaMA-3.2-3B-ins. | 55 | Not recorded | 68fa7b661e209131cea88313 |

### DeepSeek-V3

F1 Score (A): 100. Result ID: 6a211feefbc373592c1c0949.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b661e209131cea88313.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b661e209131cea88313.md>)

Recorded evaluation configuration: {"Model Category":"Instruct-LLMs","Evaluation Protocol":"10-fold cross-validation"}. This record does not assert independent reproduction.

### PRC-RoBERTa-L

F1 Score (A): 100. Result ID: 6a211feefbc373592c1c094e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b661e209131cea88313.md>)

Recorded evaluation configuration: {"Model Category":"BERT variants","Evaluation Protocol":"10-fold cross-validation"}. This record does not assert independent reproduction.

### GPT-3.5-turbo

F1 Score (A): 94. Result ID: 6a211feefbc373592c1c0948.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b661e209131cea88313.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b661e209131cea88313.md>)

Recorded evaluation configuration: {"Model Category":"Instruct-LLMs","Evaluation Protocol":"10-fold cross-validation"}. This record does not assert independent reproduction.

### Mistral-7B-ins.

F1 Score (A): 93. Result ID: 6a211feefbc373592c1c094b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b661e209131cea88313.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b661e209131cea88313.md>)

Recorded evaluation configuration: {"Model Category":"Instruct-LLMs","Evaluation Protocol":"10-fold cross-validation"}. This record does not assert independent reproduction.

### PRC-BERT-B

F1 Score (A): 93. Result ID: 6a211feefbc373592c1c094f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b661e209131cea88313.md>)

Recorded evaluation configuration: {"Model Category":"BERT variants","Evaluation Protocol":"10-fold cross-validation"}. This record does not assert independent reproduction.

### NB (DP)

F1 Score (A): 90. Result ID: 6a211feefbc373592c1c0952.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b661e209131cea88313.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63831e209131ce9ff19b.md>)

Recorded evaluation configuration: {"Model Category":"ML algorithms","Evaluation Protocol":"10-fold cross-validation","Data Processing":"Manual feature selection (DP)"}. This record does not assert independent reproduction.

### LLaMA-3.1-8B-ins.

F1 Score (A): 81. Result ID: 6a211feefbc373592c1c094c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b661e209131cea88313.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b661e209131cea88313.md>)

Recorded evaluation configuration: {"Model Category":"Instruct-LLMs","Evaluation Protocol":"10-fold cross-validation"}. This record does not assert independent reproduction.

### NoR-BERT-B

F1 Score (A): 79. Result ID: 6a211feefbc373592c1c0951.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b661e209131cea88313.md>)

Recorded evaluation configuration: {"Model Category":"BERT variants","Evaluation Protocol":"10-fold cross-validation"}. This record does not assert independent reproduction.

### NoR-BERT-L

F1 Score (A): 78. Result ID: 6a211feefbc373592c1c0950.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b661e209131cea88313.md>)

Recorded evaluation configuration: {"Model Category":"BERT variants","Evaluation Protocol":"10-fold cross-validation"}. This record does not assert independent reproduction.

### LDA (DP)

F1 Score (A): 74. Result ID: 6a211feefbc373592c1c0953.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b661e209131cea88313.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63831e209131ce9ff19b.md>)

Recorded evaluation configuration: {"Model Category":"ML algorithms","Evaluation Protocol":"10-fold cross-validation","Data Processing":"Manual feature selection (DP)"}. This record does not assert independent reproduction.

### GPT-4o

F1 Score (A): 68. Result ID: 6a211feefbc373592c1c0947.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b661e209131cea88313.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b661e209131cea88313.md>)

Recorded evaluation configuration: {"Model Category":"Instruct-LLMs","Evaluation Protocol":"10-fold cross-validation"}. This record does not assert independent reproduction.

### Gemma-3-4B

F1 Score (A): 67. Result ID: 6a211feefbc373592c1c094a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b661e209131cea88313.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b661e209131cea88313.md>)

Recorded evaluation configuration: {"Model Category":"Instruct-LLMs","Evaluation Protocol":"10-fold cross-validation"}. This record does not assert independent reproduction.

### LLaMA-3.2-3B-ins.

F1 Score (A): 55. Result ID: 6a211feefbc373592c1c094d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b661e209131cea88313.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b661e209131cea88313.md>)

Recorded evaluation configuration: {"Model Category":"Instruct-LLMs","Evaluation Protocol":"10-fold cross-validation"}. This record does not assert independent reproduction.

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