What Would Cicero Write?

Recent developments in Transformer language models now allow users to predict the probability of different sentences and to predict missing words more accurately than before. This new information and perspective can be used to form judgments on novel textual emendations and to further quantify exist...

Full description

Saved in:
Bibliographic Details
Main Author: Todd G. Cook, TGC
Format: Article
Language:deu
Published: Università degli Studi di Torino 2021-12-01
Series:Ciceroniana On Line
Online Access:https://www.ojs.unito.it/index.php/COL/article/view/6523
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849707696994910208
author Todd G. Cook, TGC
author_facet Todd G. Cook, TGC
author_sort Todd G. Cook, TGC
collection DOAJ
description Recent developments in Transformer language models now allow users to predict the probability of different sentences and to predict missing words more accurately than before. This new information and perspective can be used to form judgments on novel textual emendations and to further quantify existing historical editorial judgments. We examine the importance of analyzing an author’s corpus, and the impact of the Good-Turing theory of frequency estimation when predicting missing words. We will also outline some of the limits of what Transformer language models can do, and how to practically evaluate them.
format Article
id doaj-art-008b9ca1df304a5da7ef1a6c0257e63b
institution DOAJ
issn 2532-5299
2532-5353
language deu
publishDate 2021-12-01
publisher Università degli Studi di Torino
record_format Article
series Ciceroniana On Line
spelling doaj-art-008b9ca1df304a5da7ef1a6c0257e63b2025-08-20T03:15:51ZdeuUniversità degli Studi di TorinoCiceroniana On Line2532-52992532-53532021-12-015210.13135/2532-5353/6523What Would Cicero Write?Todd G. Cook, TGC0Classical Language Toolkit (CLTK.org)Recent developments in Transformer language models now allow users to predict the probability of different sentences and to predict missing words more accurately than before. This new information and perspective can be used to form judgments on novel textual emendations and to further quantify existing historical editorial judgments. We examine the importance of analyzing an author’s corpus, and the impact of the Good-Turing theory of frequency estimation when predicting missing words. We will also outline some of the limits of what Transformer language models can do, and how to practically evaluate them.https://www.ojs.unito.it/index.php/COL/article/view/6523
spellingShingle Todd G. Cook, TGC
What Would Cicero Write?
Ciceroniana On Line
title What Would Cicero Write?
title_full What Would Cicero Write?
title_fullStr What Would Cicero Write?
title_full_unstemmed What Would Cicero Write?
title_short What Would Cicero Write?
title_sort what would cicero write
url https://www.ojs.unito.it/index.php/COL/article/view/6523
work_keys_str_mv AT toddgcooktgc whatwouldcicerowrite