For the results to be comparable, we need to establish how Voynichese words correspond to plaintext words (and if they are words and not syllables, character combinations, etc). I think it makes sense repeating the experiment using comparable training sizes and output sizes expressed in Shannon bits
The Voynich Ninja
›
Voynich Research
›
Analysis of the text
GPT Models Fail to Find Language Structure in the Voynich Manuscript
GPT Models Fail to Find Language Structure in the Voynich Manuscript
|
GPT Models Fail to Find Language Structure in the Voynich Manuscript
|
|
« Next Oldest | Next Newest »
|
| Possibly Related Threads… | |||||
| Thread | Author | Replies | Views | Last Post | |
| Measuring Long-Range Structure in the Voynich Manuscript | quimqu | 49 | 16,223 |
28-02-2026, 09:56 AM Last Post: quimqu |
|
| How LLM models try to understand Voynichese | quimqu | 6 | 6,499 |
28-09-2025, 09:18 AM Last Post: stopsquark |
|
| Why the Voynich Manuscript Might Not Be a Real Language | quimqu | 26 | 18,409 |
26-06-2025, 05:11 AM Last Post: ReneZ |
|
| Narration structure as the tiebreaker between cipher & (synthetic) language | Anton | 12 | 22,078 |
03-02-2016, 03:40 PM Last Post: MarcoP |
|
Users browsing this thread: 1 Guest(s)