Datasets:
Release cards: no paper URL yet, takedown contact via Community tab; rerank.json names the backend of the published scores
Browse files
README.md
CHANGED
|
@@ -56,7 +56,7 @@ These are the exact **pretraining (stage-1)** pairs of the Samdei Type romanized
|
|
| 56 |
|
| 57 |
They trained the pretrained checkpoints [`samdei/samdei-type-base-pretrained`](https://huggingface.co/samdei/samdei-type-base-pretrained) and [`samdei/samdei-type-mini-pretrained`](https://huggingface.co/samdei/samdei-type-mini-pretrained), from which Samdei Type Base and Mini were fine-tuned. The fine-tuning and adaptation data (stages 2–3) is in [`samdei/samdei-type-finetuning-data`](https://huggingface.co/datasets/samdei/samdei-type-finetuning-data).
|
| 58 |
|
| 59 |
-
|
| 60 |
|
| 61 |
## Configs and files
|
| 62 |
|
|
@@ -136,7 +136,7 @@ To rebuild a mix exactly, read `recipes.json` and concatenate its files in order
|
|
| 136 |
- **Synthetic inputs.** No row is real typing; romanizations follow our four personas and one LLM.
|
| 137 |
- **Pieces, not messages.** Inputs are short (median 26 characters, at most 132).
|
| 138 |
- **Web noise.** About 1.5% of HPLT lines are misordered text extracted from PDFs.
|
| 139 |
-
- **Personal data.** Web text may contain names, phone numbers or other personal data of the original authors. Lines with URLs or e-mail-like strings were dropped, but no other personal-data filter was run.
|
| 140 |
- **LLM text.** The Gemma-written chat sentences and romanizations can contain mistakes.
|
| 141 |
|
| 142 |
## License and attribution
|
|
@@ -164,9 +164,8 @@ Rows whose `source` is `wikipedia` or `gemrom_wikipedia` carry Wikipedia's share
|
|
| 164 |
@misc{samdei-type-2026,
|
| 165 |
title = {Samdei Type: Separating What Is Written from How It Is Spelled for Romanized Khmer Transliteration},
|
| 166 |
author = {Mengsay Loem},
|
| 167 |
-
year = {2026}
|
| 168 |
-
url = {<paper-url>}
|
| 169 |
}
|
| 170 |
```
|
| 171 |
|
| 172 |
-
Contact:
|
|
|
|
| 56 |
|
| 57 |
They trained the pretrained checkpoints [`samdei/samdei-type-base-pretrained`](https://huggingface.co/samdei/samdei-type-base-pretrained) and [`samdei/samdei-type-mini-pretrained`](https://huggingface.co/samdei/samdei-type-mini-pretrained), from which Samdei Type Base and Mini were fine-tuned. The fine-tuning and adaptation data (stages 2–3) is in [`samdei/samdei-type-finetuning-data`](https://huggingface.co/datasets/samdei/samdei-type-finetuning-data).
|
| 58 |
|
| 59 |
+
Inference code: [github.com/Samdei-Labs/samdei-type](https://github.com/Samdei-Labs/samdei-type). Test set: [`samdei/samdei-type-bench`](https://huggingface.co/datasets/samdei/samdei-type-bench).
|
| 60 |
|
| 61 |
## Configs and files
|
| 62 |
|
|
|
|
| 136 |
- **Synthetic inputs.** No row is real typing; romanizations follow our four personas and one LLM.
|
| 137 |
- **Pieces, not messages.** Inputs are short (median 26 characters, at most 132).
|
| 138 |
- **Web noise.** About 1.5% of HPLT lines are misordered text extracted from PDFs.
|
| 139 |
+
- **Personal data.** Web text may contain names, phone numbers or other personal data of the original authors. Lines with URLs or e-mail-like strings were dropped, but no other personal-data filter was run. To have a row removed, open a discussion in the [Community tab](https://huggingface.co/datasets/samdei/samdei-type-pretraining-data/discussions).
|
| 140 |
- **LLM text.** The Gemma-written chat sentences and romanizations can contain mistakes.
|
| 141 |
|
| 142 |
## License and attribution
|
|
|
|
| 164 |
@misc{samdei-type-2026,
|
| 165 |
title = {Samdei Type: Separating What Is Written from How It Is Spelled for Romanized Khmer Transliteration},
|
| 166 |
author = {Mengsay Loem},
|
| 167 |
+
year = {2026}
|
|
|
|
| 168 |
}
|
| 169 |
```
|
| 170 |
|
| 171 |
+
Contact: open a discussion in this repository's [Community tab](https://huggingface.co/datasets/samdei/samdei-type-pretraining-data/discussions).
|