Stoicheia: Character-Level Masked Diffusion for Ancient Greek Textual Restoration, Parsing, and Metrical Scansion
Eric Cullhed, Albin Thörn Cleland
cs.CL
Submitted: 2026-08-07
Updated: 2026-08-10
Comments: 12 pages, 7 tables. Models, datasets and code released: https://huggingface.co/collections/Ericu950/stoicheia-6a6fbf9800c82d93020a7ceb and https://github.com/ericu9500/stoicheia
Code: https://github.com/ericu9500/stoicheia
License: http://creativecommons.org/licenses/by/4.0/
The gist: We introduce Stoicheia, a 405M-parameter character-level masked-diffusion encoder for Ancient Greek whose input factors into five aligned, independently maskable planes: letters, word and sentence
Terminology
Abstract
We introduce Stoicheia, a 405M-parameter character-level masked-diffusion encoder for Ancient Greek whose input factors into five aligned, independently maskable planes: letters, word and sentence boundaries, diacritics, capitalization, and punctuation. A single backbone can therefore restore lacunae, re-segment, accentuate, and punctuate unspaced text without task-specific retokenization. We pretrain it on an open, revision-pinned corpus of 380M words and release eleven checkpoints: ten rotated, decontaminated folds, guaranteeing that for any given literary passage at least one released model has never seen its text, and one with no exposure to documentary texts. Three experiments - reconstruction of damaged inscriptions and papyri, morphosyntactic tagging and dependency parsing, and macronization with metrical scansion - each carry a matched random-initialization control, isolating what character-level diffusion pretraining contributes: 5.6 CER points on inscription reconstruction, 12.9 LAS on parsing, and 6.0 points of balanced accuracy on macronization. On Ithaca's own test split, with identical frozen samples and strict scoring, Stoicheia reduces character error relative to both prior state-of-the-art systems, from 24.6 (Ithaca) and 23.5 (its 2025 Aeneas-framework successor) to 15.5, and raises top-1 accuracy from 63.0 and 64.0 to 74.5.
Sources
- Quantifying Memorization Across Neural Language Models
- Opera Graeca Adnotata: Building a 34M+ Token Multilayer Corpus for Ancient Greek
- A State-of-the-Art Morphosyntactic Parser and Lemmatizer for Ancient Greek
- Structure-Aware Text Recognition for Ancient Greek Critical Editions
- Automatic Annotation of Ancient Greek Vowel Length
- GPT-4o System Card
- Instruct-Tuning Pretrained Causal Language Models for Ancient Greek Papyrology and Epigraphy
Related papers
- Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
- Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
- Subliminal Steering: Stronger Encoding of Hidden Signals
- MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
- The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
- Untangling the Mechanisms of Misleading Context in Medical Question Answering