1602353775wzj's picture
Expand dataset card for review: source, size, splits, construction, license, metrics, citations
86bcedf verified
|
Raw
History Blame Contribute Delete
11.6 kB
---
license: cc-by-nc-4.0
task_categories:
- translation
- sentence-similarity
task_ids:
- semantic-similarity-scoring
- document-retrieval
language:
- en
- de
- nl
- it
- pt
- ru
- es
- ro
- sv
- uk
multilinguality:
- multilingual
- translation
size_categories:
- 100K<n<1M
pretty_name: PoetryTranslationEMNLP2021
tags:
- poetry
- bitext-mining
- retrieval
- parallel-corpus
- multilingual
- mteb
- poetrymteb
- sentence-similarity
configs:
- config_name: deu-eng
data_files:
- split: train
path: deu-eng/train-*
- split: validation
path: deu-eng/validation-*
- split: test
path: deu-eng/test-*
default: true
- config_name: ita-eng
data_files:
- split: train
path: ita-eng/train-*
- split: validation
path: ita-eng/validation-*
- split: test
path: ita-eng/test-*
- config_name: nld-eng
data_files:
- split: train
path: nld-eng/train-*
- split: validation
path: nld-eng/validation-*
- split: test
path: nld-eng/test-*
- config_name: por-eng
data_files:
- split: train
path: por-eng/train-*
- split: validation
path: por-eng/validation-*
- split: test
path: por-eng/test-*
- config_name: ron-eng
data_files:
- split: test
path: ron-eng/test-*
- config_name: rus-eng
data_files:
- split: train
path: rus-eng/train-*
- split: validation
path: rus-eng/validation-*
- split: test
path: rus-eng/test-*
- config_name: spa-eng
data_files:
- split: train
path: spa-eng/train-*
- split: validation
path: spa-eng/validation-*
- split: test
path: spa-eng/test-*
- config_name: swe-eng
data_files:
- split: test
path: swe-eng/test-*
- config_name: ukr-eng
data_files:
- split: test
path: ukr-eng/test-*
dataset_info:
- config_name: deu-eng
features:
- name: id
dtype: string
- name: source_text
dtype: string
- name: target_text
dtype: string
splits:
- name: train
num_bytes: 1880402
num_examples: 16999
- name: validation
num_bytes: 115087
num_examples: 1051
- name: test
num_bytes: 119503
num_examples: 1296
download_size: 1680748
dataset_size: 2114992
- config_name: ita-eng
features:
- name: id
dtype: string
- name: source_text
dtype: string
- name: target_text
dtype: string
splits:
- name: train
num_bytes: 3833784
num_examples: 34525
- name: validation
num_bytes: 219099
num_examples: 1998
- name: test
num_bytes: 31173
num_examples: 315
download_size: 3188013
dataset_size: 4084056
- config_name: nld-eng
features:
- name: id
dtype: string
- name: source_text
dtype: string
- name: target_text
dtype: string
splits:
- name: train
num_bytes: 2506045
num_examples: 23376
- name: validation
num_bytes: 106021
num_examples: 1000
- name: test
num_bytes: 16115
num_examples: 159
download_size: 2091864
dataset_size: 2628181
- config_name: por-eng
features:
- name: id
dtype: string
- name: source_text
dtype: string
- name: target_text
dtype: string
splits:
- name: train
num_bytes: 1556212
num_examples: 15200
- name: validation
num_bytes: 77272
num_examples: 700
- name: test
num_bytes: 14576
num_examples: 140
download_size: 1272443
dataset_size: 1648060
- config_name: ron-eng
features:
- name: id
dtype: string
- name: source_text
dtype: string
- name: target_text
dtype: string
splits:
- name: test
num_bytes: 11418
num_examples: 100
download_size: 10703
dataset_size: 11418
- config_name: rus-eng
features:
- name: id
dtype: string
- name: source_text
dtype: string
- name: target_text
dtype: string
splits:
- name: train
num_bytes: 5892615
num_examples: 50001
- name: validation
num_bytes: 491361
num_examples: 4186
- name: test
num_bytes: 72561
num_examples: 548
download_size: 4584359
dataset_size: 6456537
- config_name: spa-eng
features:
- name: id
dtype: string
- name: source_text
dtype: string
- name: target_text
dtype: string
splits:
- name: train
num_bytes: 3825733
num_examples: 37744
- name: validation
num_bytes: 200699
num_examples: 2059
- name: test
num_bytes: 53043
num_examples: 536
download_size: 3280916
dataset_size: 4079475
- config_name: swe-eng
features:
- name: id
dtype: string
- name: source_text
dtype: string
- name: target_text
dtype: string
splits:
- name: test
num_bytes: 9407
num_examples: 96
download_size: 9435
dataset_size: 9407
- config_name: ukr-eng
features:
- name: id
dtype: string
- name: source_text
dtype: string
- name: target_text
dtype: string
splits:
- name: test
num_bytes: 11031
num_examples: 99
download_size: 10116
dataset_size: 11031
---
# PoetryTranslationEMNLP2021
Poetry-to-English parallel poetry bitext, reformatted for **PoetryMTEB** Bitext Mining / cross-lingual retrieval evaluation.
This Hub release is a structured redistribution of the parallel poetry corpus released with the EMNLP 2021 paper *Don't Go Far Off: An Empirical Study on Neural Poetry Translation* ([Chakrabarty et al., 2021](https://aclanthology.org/2021.emnlp-main.577/); source repository: [tuhinjubcse/PoetryTranslationEMNLP2021](https://github.com/tuhinjubcse/PoetryTranslationEMNLP2021)).
---
## Dataset Card
| Item | Description |
|------|-------------|
| **Source** | [PoetryTranslationEMNLP2021](https://github.com/tuhinjubcse/PoetryTranslationEMNLP2021) (`corpus_v2.1` + `testdatawithtranslations`); paper [Chakrabarty et al., EMNLP 2021](https://aclanthology.org/2021.emnlp-main.577/) |
| **Languages** | Source → English: German (`deu`), Dutch (`nld`), Italian (`ita`), Portuguese (`por`), Russian (`rus`), Spanish (`spa`); **test-only**: Romanian (`ron`), Swedish (`swe`), Ukrainian (`ukr`) |
| **Size** | ~192K aligned pairs across 9 configs (see per-config table below) |
| **Label / field type** | Parallel bitext (no classification labels). Fields: `id` (string), `source_text` (string), `target_text` (string, English) |
| **Splits** | `train` / `validation` (6 language pairs from `corpus_v2.1`); `test` (all 9 pairs from `testdatawithtranslations`) |
| **Construction** | See [Construction method](#construction-method) |
| **License** | [CC BY-NC 4.0](https://creativecommons.org/licenses/by-nc/4.0/) |
| **Evaluation metrics** | Bitext mining / retrieval: **Recall@k**, **nDCG@k** (PoetryMTEB). Original MT study also reports **BLEU**, **BERTScore**, **COMET**, and human faithfulness / poetic-style ratings |
---
## Features
| Field | Type | Description |
|-------|------|-------------|
| `id` | `string` | Unique pair identifier |
| `source_text` | `string` | Non-English source poetry text |
| `target_text` | `string` | English target / gold translation |
Aligned with PoetryMTEB `MultilingualPoetryBitextMining` schema for drop-in bitext mining evaluation.
---
## Configs, languages & size
Each Hugging Face **config** is one language pair `{src}-eng` (ISO 639-3).
| Config | Source lang | Train | Validation | Test | Total |
|--------|-------------|------:|-----------:|-----:|------:|
| `deu-eng` | German | 16,999 | 1,051 | 1,296 | 19,346 |
| `ita-eng` | Italian | 34,525 | 1,998 | 315 | 36,838 |
| `nld-eng` | Dutch | 23,376 | 1,000 | 159 | 24,535 |
| `por-eng` | Portuguese | 15,200 | 700 | 140 | 16,040 |
| `rus-eng` | Russian | 50,001 | 4,186 | 548 | 54,735 |
| `spa-eng` | Spanish | 37,744 | 2,059 | 536 | 40,339 |
| `ron-eng` | Romanian | — | — | 100 | 100 |
| `swe-eng` | Swedish | — | — | 96 | 96 |
| `ukr-eng` | Ukrainian | — | — | 99 | 99 |
| **All** | | **177,845** | **10,994** | **3,289** | **192,128** |
---
## Construction method
1. **Upstream corpus**
Parallel poetry translations collected and released by Chakrabarty et al. (EMNLP 2021):
- `corpus_v2.1`: sentence-level `{lang}train.json` / `{lang}valid.json` for 6 languages → English
- `testdatawithtranslations`: poem-level source (`test*`) and gold English (`gold*`) for 6 + 3 additional languages
2. **This release (PoetryMTEB packaging)**
- **train / validation**: read sentence pairs from `corpus_v2.1`, map fields to `source_text` / `target_text`, assign stable `id`s, write Parquet shards per config.
- **test**: for each poem, split source and gold into non-empty lines; align lines within the poem (truncate to the shorter side if lengths differ); emit one bitext row per aligned line.
- Schema unified with PoetryMTEB Bitext Mining (`id`, `source_text`, `target_text`).
We do **not** redistribute model-generated candidates from the original repo; only human gold references are used as `target_text` on the test split.
---
## Evaluation metrics
**Primary (PoetryMTEB Bitext Mining / retrieval):**
| Metric | Role |
|--------|------|
| Recall@k | Fraction of queries whose gold translation is retrieved in top-k |
| nDCG@k | Ranking quality against gold bitext partners |
**Reported in the original EMNLP 2021 study (MT setting):** BLEU, BERTScore, COMET, plus human evaluation of meaning faithfulness and poetic style ([paper](https://aclanthology.org/2021.emnlp-main.577/)).
---
## How to load
```python
from datasets import load_dataset
ds = load_dataset("PoetryMTEB/PoetryTranslationEMNLP2021", "deu-eng")
print(ds)
print(ds["test"][0])
# {'id': '...', 'source_text': '...', 'target_text': '...'}
```
Available configs: `deu-eng`, `ita-eng`, `nld-eng`, `por-eng`, `rus-eng`, `spa-eng`, `ron-eng`, `swe-eng`, `ukr-eng`.
---
## License
Distributed under [Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)](https://creativecommons.org/licenses/by-nc/4.0/).
Please also respect any terms associated with the [original GitHub release](https://github.com/tuhinjubcse/PoetryTranslationEMNLP2021).
---
## Citation
If you use this dataset, please cite the original paper and, when relevant, this PoetryMTEB redistribution.
**Paper (required):**
```bibtex
@inproceedings{chakrabarty-etal-2021-dont,
title = "Don{'}t Go Far Off: An Empirical Study on Neural Poetry Translation",
author = "Chakrabarty, Tuhin and
Saakyan, Arkadiy and
Muresan, Smaranda",
editor = "Moens, Marie-Francine and
Huang, Xuanjing and
Specia, Lucia and
Yih, Scott Wen-tau",
booktitle = "Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2021",
address = "Online and Punta Cana, Dominican Republic",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.emnlp-main.577/",
doi = "10.18653/v1/2021.emnlp-main.577",
pages = "7253--7265"
}
```
**Dataset source repository:**
- GitHub: [https://github.com/tuhinjubcse/PoetryTranslationEMNLP2021](https://github.com/tuhinjubcse/PoetryTranslationEMNLP2021)
- ACL Anthology: [https://aclanthology.org/2021.emnlp-main.577/](https://aclanthology.org/2021.emnlp-main.577/)
- This Hub dataset: `PoetryMTEB/PoetryTranslationEMNLP2021`