# Poetry Translation on English-Chinese Poetry Translation Human Evaluation Dataset (test)

Source: [Poetry Translation on English-Chinese Poetry Translation Human Evaluation Dataset (test)](<https://www.sota2.com/research/sota/poetry-translation-on-english-chinese-poetry-translation-human-evaluation-dataset-test>)

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

Data updated: 2026-02-26T05:54:30.411Z

Human evaluation results for Machine Translation performance specifically on English-Chinese poetry, assessed using metrics such as OI, Sim, Fide, and Poet 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 | Poetry Translation |
| Dataset (including any stated split) | English-Chinese Poetry Translation Human Evaluation Dataset (test) |
| Metric | Overall Impression (OI) |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| 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

- [Overall Impression (OI)](<https://www.sota2.com/research/sota/poetry-translation-on-english-chinese-poetry-translation-human-evaluation-dataset-test.md?metric=Overall%20Impression%20(OI)>): Higher is better; unit: SCALAR.

- [Similarity (Sim)](<https://www.sota2.com/research/sota/poetry-translation-on-english-chinese-poetry-translation-human-evaluation-dataset-test.md?metric=Similarity%20(Sim)>): Higher is better; unit: SCALAR.

- [Fidelity (Fide)](<https://www.sota2.com/research/sota/poetry-translation-on-english-chinese-poetry-translation-human-evaluation-dataset-test.md?metric=Fidelity%20(Fide)>): Higher is better; unit: SCALAR.

- [Line Quality (Line)](<https://www.sota2.com/research/sota/poetry-translation-on-english-chinese-poetry-translation-human-evaluation-dataset-test.md?metric=Line%20Quality%20(Line)>): Higher is better; unit: SCALAR.

- [Mean Score (Mean)](<https://www.sota2.com/research/sota/poetry-translation-on-english-chinese-poetry-translation-human-evaluation-dataset-test.md?metric=Mean%20Score%20(Mean)>): Higher is better; unit: SCALAR.

- [Poetic Quality (Poet)](<https://www.sota2.com/research/sota/poetry-translation-on-english-chinese-poetry-translation-human-evaluation-dataset-test.md?metric=Poetic%20Quality%20(Poet)>): Higher is better; unit: SCALAR.

- [Accuracy (Acc)](<https://www.sota2.com/research/sota/poetry-translation-on-english-chinese-poetry-translation-human-evaluation-dataset-test.md?metric=Accuracy%20(Acc)>): Higher is better; unit: SCALAR.

- [Error Rate (Erro)](<https://www.sota2.com/research/sota/poetry-translation-on-english-chinese-poetry-translation-human-evaluation-dataset-test.md?metric=Error%20Rate%20(Erro)>): Lower is better; unit: SCALAR.

## Recorded results

| Method | Overall Impression (OI) | Result date | Report paper |
| --- | --- | --- | --- |
| EAPMT-4.0 | 4 | Not recorded | 68fa73871e209131cea5cf77 |
| EAPMT-3.5 | 3.82 | Not recorded | 68fa73871e209131cea5cf77 |
| GPT4-Best | 3.77 | Not recorded | 68fa73871e209131cea5cf77 |
| GPT3.5-Best | 3.58 | Not recorded | 68fa73871e209131cea5cf77 |
| Online System | 3 | Not recorded | 68fa73871e209131cea5cf77 |

### EAPMT-4.0

Overall Impression (OI): 4. Result ID: 6992e6d73a6c314849124424.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73871e209131cea5cf77.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73871e209131cea5cf77.md>)

[Implementation](<https://github.com/NLP2CT/Poetry-Translation>)

Recorded evaluation configuration: {"Backbone":"GPT-4.0","Method":"Explanation-Assisted Poetry Machine Translation"}. This record does not assert independent reproduction.

### EAPMT-3.5

Overall Impression (OI): 3.82. Result ID: 6992e6d73a6c314849124422.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73871e209131cea5cf77.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73871e209131cea5cf77.md>)

[Implementation](<https://github.com/NLP2CT/Poetry-Translation>)

Recorded evaluation configuration: {"Backbone":"GPT-3.5","Method":"Explanation-Assisted Poetry Machine Translation"}. This record does not assert independent reproduction.

### GPT4-Best

Overall Impression (OI): 3.77. Result ID: 6992e6d73a6c314849124423.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73871e209131cea5cf77.md>)

Recorded evaluation configuration: {"Backbone":"GPT-4","Configuration":"Best prompt"}. This record does not assert independent reproduction.

### GPT3.5-Best

Overall Impression (OI): 3.58. Result ID: 6992e6d73a6c314849124421.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73871e209131cea5cf77.md>)

Recorded evaluation configuration: {"Backbone":"GPT-3.5","Configuration":"Best prompt"}. This record does not assert independent reproduction.

### Online System

Overall Impression (OI): 3. Result ID: 6992e6d73a6c314849124420.

[Reporting paper](<https://www.sota2.com/research/paper/68fa73871e209131cea5cf77.md>)

Recorded evaluation configuration: {"Type":"Generic Online Translation"}. This record does not assert independent reproduction.

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