Run · Voynich Manuscript
T43 generation 2: Gromov-Wasserstein word alignment with a homophone-free word code (N 150, epsilon 0.01); method dropped
gw_t43.py now takes the control setting from the SETTINGS environment variable (otherwise unchanged from generation 1).
Result
| Run | Measured | Result | Against | Decision | Date |
|---|---|---|---|---|---|
| Run | Held-out word accuracy of the coupling's word map (bar 0.5) | word-spacing control 0.039; stream-spacing control 0.0007 | Generation 1: 0.003 | Fails its control in a second generation, with the easiest case it is built for (one plaintext word per written word, no homophones, 400-word nomenclator, Herbal B's NPE setting with word spacing). Dropped per the brief. A word-level map can't be detected by co-occurrence at this text length, so the absence of one on the manuscript would say nothing; the method is not run on it. |
What it used and found
Latin genre texts
DATDataset
Parameters
N = 150, window = 2, epsilon = 0.01, control_setting = {'K': 400, 'h_c': 1, 'h_v': 1, 'lam': 0.5, 'Ssyl': 0, 'abbrev': 0, 'p_copy': 0, 'p_stay': 0.98, 'window': 20, 'novowel': 0, 'p_break': 0.9, 'spacing': 'word (control) / stream', 'simplify': 1, 'positional': 0}Cite it
Mysteries Department. (2026). T43 generation 2: Gromov-Wasserstein word alignment with a homophone-free word code (N 150, epsilon 0.01); method dropped [Run RUN0DCF9GEQF284]. https://mysteriesdept.com/r/RUN0DCF9GEQF284
@misc{mysteriesdept_run0dcf9geqf284,
title = {T43 generation 2: Gromov-Wasserstein word alignment with a homophone-free word code (N 150, epsilon 0.01); method dropped},
author = {{Mysteries Department}},
year = {2026},
note = {Run, catalog key RUN0DCF9GEQF284},
url = {https://mysteriesdept.com/r/RUN0DCF9GEQF284}
}