| --- |
| language: grc |
| license: gpl-3.0 |
| tags: |
| - ancient-greek |
| - vowel-length |
| - macronization |
| - token-classification |
| - char-level |
| --- |
| |
| # grc-macronizer-char: a small char-level model for Ancient Greek vowel-length annotation |
|
|
| This model predicts the phonemic length (long/short/undetermined) of the three |
| Ancient Greek "dichrona" -- alpha (α), iota (ι), and upsilon (υ) -- whose length |
| is not disambiguated by the standard Greek script. It was trained on Ancient |
| Greek text automatically macronized by the rule-based |
| [grc-macronizer](https://github.com/Urdatorn/grc-macronizer), applied to the |
| [Opera Graeca Adnotata](https://doi.org/10.5281/zenodo.14206061) corpus |
| (Celano 2024), and is intended as a complementary, corpus-general alternative |
| to that rule-based system. |
|
|
| ## Architecture |
|
|
| A small transformer encoder (~0.9M parameters) operating over three character-level |
| "planes" per input position: |
|
|
| - **plane 1 (letter)**: which of the 24 Greek letters (final/medial sigma |
| folded together), or space, or "other" (punctuation/digits/foreign chars) |
| - **plane 2 (diacritic)**: the combination of accent/breathing/diaeresis marks |
| on that letter, if any (a small vocabulary fit from the training data) |
| - **target (macron)**: none / short / long -- predicted only at positions |
| that are genuine ambiguous dichrona (diphthong members and circumflexed |
| vowels are excluded, matching the rule-based system's own definition) |
|
|
| During training, the diacritic plane is randomly masked (accents stripped) |
| for a fraction of characters, so the model also learns to macronize |
| unaccented input. |
|
|
| ## Training data |
|
|
| Training labels come directly from the rule-based macronizer's own output: |
| wherever it marks a dichronon (it does not guess -- see the |
| [macronizer paper](https://github.com/Urdatorn/grc-macronizer)), that becomes |
| a training label; dichrona it left unmarked are excluded from the loss |
| entirely (never treated as a negative "short" or "long" example), since the |
| rule-based system is known to be incomplete rather than wrong where it commits. |
|
|
| ## Evaluation |
|
|
| Evaluated on the full [Norma Syllabarum Graecarum](https://huggingface.co/datasets/Urdatorn/norma) |
| benchmark, spanning all 17 manually-annotated works (prose and verse; 2,957 |
| gold-annotated positions). Scored with the identical harness |
| (`macron_model/eval_norma.py`) against the bundled rule-based macronizer, |
| counting a left-unmarked dichronon as an implicit "short" guess for both |
| systems (`default_short_accuracy` below) so abstention isn't unfairly |
| penalized either way: |
|
|
| | system | accuracy (unmarked defaults to short) | raw accuracy | unmarked / total | |
| |---|---|---|---| |
| | rule-based grc-macronizer | 86.84% | 53.77% | 1,300 / 2,957 | |
| | **this model** | **89.89%** | **89.79%** | **9 / 2,957** | |
|
|
| The model outperforms the rule-based system on the fair metric while resolving |
| essentially every position (99.7% coverage vs. 56%), since it learned to |
| generalize past the cases the rule-based system couldn't commit to. The margin |
| is statistically significant (McNemar's test, continuity-corrected, |
| χ²=10.04, p=0.0015). |
|
|
| Two additional training runs with different random seeds confirm this is not |
| a lucky draw: defaults-to-short accuracy across three seeds averages |
| 89.36±0.78% (88.47% / 89.72% / **89.89%**, this checkpoint), with every |
| seed beating the rule-based teacher. |
|
|
| ## Usage |
|
|
| ```python |
| import sys |
| sys.path.insert(0, "path/to/grc-macronizer/macron_model") # for predict.py |
| from predict import MacronPredictor |
| |
| predictor = MacronPredictor("path/to/downloaded/checkpoint") |
| macronized = predictor.macronize("ανθρωπος ανηρ") |
| # -> "α^νθρωπος α_νηρ" |
| ``` |
|
|
| `MacronPredictor` expects the checkpoint directory (containing |
| `config.json`/`model.safetensors`) plus a sibling or parent `diacritic_vocab.json` |
| (see `predict.py` for the exact lookup logic). The `predict.py`/`tokenizer.py` |
| source lives in [grc-macronizer/macron_model](https://github.com/Urdatorn/grc-macronizer/tree/oga-macronization-improvements/macron_model). |
|
|
| ## Citation |
|
|
| If you use this model, please cite: |
|
|
| > Thörn Cleland, Albin and Eric Cullhed (forthcoming). Automatic Annotation of Ancient Greek Vowel Length. |
|
|
| ## License |
|
|
| GNU GPL v3, matching [grc-macronizer](https://github.com/Urdatorn/grc-macronizer) |
| (the training-data generator this model's code ships alongside). |
|
|
