<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=Windows-1252">
</head>
<body>
<div style="direction: ltr; font-family: Aptos, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<i>Anteeksi päällekkäispostitukset / Apologies for cross-posting.</i></div>
<div style="direction: ltr; font-family: Aptos, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<i><br>
</i></div>
<div style="direction: ltr; font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<span style="line-height: 1.284; text-transform: none;">Dear all,</span></div>
<div style="direction: ltr; font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<span style="line-height: 1.284; text-transform: none;"><br>
</span></div>
<div style="direction: ltr; text-align: left; text-indent: 0px; line-height: 1.284; text-transform: none; margin: 0cm 0cm 8pt; font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
for your information: the Classics Research Seminar has a special guest lecture by Dr. Alessandro Papini (University of Venezia Ca’ Foscari) titled
<b>“When the future meets the past: AI-assisted normalisation of historical Late Latin (ChLA) texts with AI_Stripertus”</b> on Thursday, March 19, at 4–6 pm. The lecture room is Metsätalo 4 (B214). All welcome; please find Alessandro’s abstract below.</div>
<div style="direction: ltr; text-align: left; text-indent: 0px; line-height: 1.284; text-transform: none; margin: 0cm 0cm 8pt; font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 12pt;">
<span style="color: rgb(0, 0, 0);">In addition, Papini presents a lecture on Tuesday, March 17, 6–8 pm (main building, U3029) with the title
<b>"Sociolinguistics, statistics, and epigraphy: Vowel mergers in Latin inscriptions from Rome (120 BCE – 600 CE)”</b> at an event organised together by the Association of Classical Philology and the study track of Classical languages (</span><span style="color: rgb(70, 120, 134);"><u><a href="https://www.kfy.fi/tapahtumat/" id="OWAefc52eb4-200d-58c4-ae52-7b6d0c490210" class="x_OWAAutoLink" title="https://www.kfy.fi/tapahtumat/" data-auth="NotApplicable" data-outlook-id="2b70f4dc-2cae-44e0-8f62-f8e75e61cae3" style="color: rgb(70, 120, 134);">https://www.kfy.fi/tapahtumat/</a></u></span><span style="color: rgb(0, 0, 0);">).
</span></div>
<div style="direction: ltr; text-align: left; text-indent: 0px; line-height: 1.284; text-transform: none; margin: 0cm 0cm 8pt; font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
Best wishes,</div>
<div style="direction: ltr; text-align: left; text-indent: 0px; line-height: 1.284; text-transform: none; margin: 0cm 0cm 8pt; font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
Timo Korkiakangas and Marja Vierros</div>
<div style="direction: ltr; text-align: left; text-indent: 0px; line-height: 1.284; text-transform: none; margin: 0cm 0cm 8pt; font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
</div>
<div style="direction: ltr; text-align: left; text-indent: 0px; line-height: 1.284; text-transform: none; margin: 0cm 0cm 8pt; font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 11pt; color: rgb(0, 0, 0);">
Abstract</div>
<div style="direction: ltr; text-align: left; text-indent: 0px; line-height: 1.284; text-transform: none; margin: 0cm 0cm 8pt; font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 11pt; color: rgb(0, 0, 0);">
When the future meets the past: AI-assisted normalisation of historical Late Latin (ChLA) texts with AI_Stripertus</div>
<div style="direction: ltr; text-align: left; text-indent: 0px; line-height: 1.284; text-transform: none; margin: 0cm 0cm 8pt; font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 11pt;">
<span style="color: rgb(0, 0, 0);">Following the publication of <i>Attention is All You Need
</i>in 2017 and the affirmation of the Transformer architecture, there has been a rapid increase in the adoption of generative AI and large language models (LLMs) in scientific research (Haber
<i>et al.</i> 2025), including the study of Latin and Greek heritage (Assel <i>et al.</i> 2022; 2025). However, thus far, AI’s capabilities have primarily been used to help scholars date and establish the provenance of ancient texts (mostly inscriptions) and
fill gaps in historical documents, as part of large-scale projects such as </span>
<span style="color: rgb(70, 120, 134);"><i><u><a href="https://predictingthepast.com/ithaca" id="OWAcc93a72c-fba4-b724-6276-698ad9eeeb9a" class="x_OWAAutoLink" title="https://predictingthepast.com/ithaca" data-auth="NotApplicable" data-outlook-id="d1b21382-44fd-4832-90b2-ef6cb48fc7ce" style="color: rgb(70, 120, 134);">Ithaca</a></u></i></span><span style="color: rgb(0, 0, 0);"> and
</span><span style="color: rgb(70, 120, 134);"><i><u><a href="https://predictingthepast.com/aeneas" id="OWA9b02209b-b9f6-0a96-0295-a51063f53c03" class="x_OWAAutoLink" title="https://predictingthepast.com/aeneas" data-auth="NotApplicable" data-outlook-id="a7abbc21-6a9f-4410-a4b8-7463a6dcae0c" style="color: rgb(70, 120, 134);">Aeneas</a></u></i></span><span style="color: rgb(0, 0, 0);"> (Tupman
2025). This paper aims to demonstrate how Transformers can also fulfil complex natural language processing (NLP) tasks for closed-corpus languages such as Latin. To this end, the paper will discuss
<i>AI_Stripertus</i>, an NLP tool leveraging agentic AI to achieve text normalisation/style transfer (<i>i.e.,
</i>“translation” from “Vulgar” to “Classical” Latin; cf. Herman 2000 and Clackson 2011) in the parchment and papyrus documents of the
<i>Chartae Latinae Antiquiores </i>(Bruckner and Marichal 1954-1985). Specifically, the paper will illustrate how the flexibility of Transformers can satisfactorily achieve this objective with limited “training” and evaluation data, contrasting with more traditional
NLP frameworks. Furthermore, the paper will discuss the contemporary linguistic/AI engineering techniques necessary for text normalisation in Latin, including, the definition of a “gold standard” for evaluation (benchmark), the creation of a knowledge base
for the model (Retrieval Augmented Generation), and the implementation of effective prompting methods (structured prompt with few shots). Finally, the paper will compare the performance of different models for the same normalisation task (reasoning vs. non-reasoning),
as well as discussing the future developments and applications of <i>AI_Stripertus
</i>in the context the ERC project <i>Digital Latin Dialectology </i>(</span><span style="color: rgb(70, 120, 134);"><u><a href="https://pric.unive.it/projects/diladi/home" id="OWA04bb0726-f8c3-2f7f-a915-8592750c4baf" class="x_OWAAutoLink" title="https://pric.unive.it/projects/diladi/home" data-auth="NotApplicable" data-outlook-id="ca6c476f-d6bb-4907-8cd2-7c243ae3b707" style="color: rgb(70, 120, 134);">DiLaDi</a></u></span><span style="color: rgb(0, 0, 0);">)
at Ca’Foscari University of Venice.</span></div>
<div style="direction: ltr; text-align: left; text-indent: 0px; line-height: 1.284; text-transform: none; margin: 0cm 0cm 8pt; font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 11pt; color: rgb(0, 0, 0);">
<b>Bibliography</b></div>
<div style="direction: ltr; text-align: left; text-indent: 0px; line-height: 1.284; text-transform: none; margin: 0cm 0cm 8pt; font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 11pt; color: rgb(0, 0, 0);">
Assael, Y. <i>et al.</i> (2022). <i>Restoring and attributing ancient texts using deep neural networks</i>. Nature, 603, 280-299.</div>
<div style="direction: ltr; text-align: left; text-indent: 0px; line-height: 1.284; text-transform: none; margin: 0cm 0cm 8pt; font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 11pt; color: rgb(0, 0, 0);">
Assael, Y. <i>et al.</i> (2025). <i>Contextualizing ancient texts with generative neural networks</i>. Nature, 645, 141-164.</div>
<div style="direction: ltr; text-align: left; text-indent: 0px; line-height: 1.284; text-transform: none; margin: 0cm 0cm 8pt; font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 11pt; color: rgb(0, 0, 0);">
Bruckner, A. and Marichal, R. (1954-1985). <i>Chartae Latinae Antiquiores. Facsimile-edition of the Latin charters prior to the Ninth Century</i>. Olten.</div>
<div style="direction: ltr; text-align: left; text-indent: 0px; line-height: 1.284; text-transform: none; margin: 0cm 0cm 8pt; font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 11pt; color: rgb(0, 0, 0);">
Clackson, J. (2011). <i>Classical Latin. </i>In, Clackson, J. (ed.), <i>A companion to the Latin language</i>. Chichester, West Sussex and Malden, MA, pp. 236-256.</div>
<div style="direction: ltr; text-align: left; text-indent: 0px; line-height: 1.284; text-transform: none; margin: 0cm 0cm 8pt; font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 11pt; color: rgb(0, 0, 0);">
Haber, E. <i>et al. </i>(2025). Generative AI for Research. In, Haber, E. <i>et al.
</i>(eds.), <i>Using AI in Academic Writing and Research</i>. Cham, pp. 27-38.</div>
<div style="direction: ltr; text-align: left; text-indent: 0px; line-height: 1.284; text-transform: none; margin: 0cm 0cm 8pt; font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 11pt; color: rgb(0, 0, 0);">
Herman, J. (2000). <i>Vulgar Latin</i>. University Park, Pa.</div>
<div style="direction: ltr; text-align: left; text-indent: 0px; line-height: 1.284; text-transform: none; margin: 0cm 0cm 8pt; font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 11pt; color: rgb(0, 0, 0);">
Tupman, C. (2025). <i>AI for Latin inscriptions supplies missing text and predicts date and location</i>. Nature, 645, 43-44.</div>
<div style="color: rgb(0, 0, 0);"><span style="font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 11pt; line-height: 1.284; text-transform: none;">Vaswani, A.
<i>et al. </i>(2017). <i>Attention is all you need</i>. Advances in neural information processing systems, 30.</span><span style="font-family: Aptos, Arial, Helvetica, sans-serif; font-size: 12pt;"><br>
<br>
</span></div>
<div style="direction: ltr; font-family: Aptos, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<i><br>
</i></div>
</body>
</html>