• Anna Folta-Rusin (Dr) Jagiellonian University

Jagiellonian University

The article describes an experiment conducted during a class entitled Representation of a Material Object at the turn of December and January 2025/26 with the participation of students of electronic information processing. The experiment aimed to test how large language models cope with the interpretation of metaphors, which are semantic stylistic devices, and their transposition into images. The research material used in the experiment consisted of poetic and critical ekphrases of works of art. Ekphrases were chosen because, historically, the primary function of this genre was to evoke image-like effects in the mind of the recipient, known as imaginings or visualizations. The experiment was preceded by three stages. The first stage involved reviewing scientific works devoted to identifying the stylistic specificity of the ekphrasis genre. This enabled the identification of stylistic devices with polysemic properties that are characteristic of this text genre. The second stage involved entering a prompt into the image generator in the form of a selected ekphrasis containing ambiguous semantic means. Then, a group of students attempted to draw conclusions based on images obtained from the artificial intelligence model. The third stage involves attempts to change the prompt in order to obtain meanings similar to the interpretative conclusions that a human being might draw from a polysemous text. The experiment made it possible to identify and present the possibilities of interpretation and then the visualization of metaphors by artificial intelligence. A presentation at the conference will analyse the process and results of this experiment. This raises the question of whether AI-based image generators will transform the long-standing competition between words and images.

• Michael J Sullivan (Dr) University of Oxford

University of Oxford

This paper reveals findings and practices from Recovery of Literary Manuscripts, an Arts-STEM collaboration to restore lost lines of anglophone literature that had been erased and environmentally damaged (www.recovery-of-literary-manuscripts.net). In this project, we apply and advance techniques of multispectral analysis for the study of nineteenth-century manuscripts, with particular attention to the digital processes and visualization methods that can enable the restoration of lost cultural text. Digitally stripping away deletions in manuscripts has transformative consequences for our relations with literary archives, turning previously examined collections into untapped repositories of lost literary lines now newly visible. From the revised manuscripts of Alfred Tennyson to water-damaged works of the Shelley Circle, the project’s methods have yielded new insights into factors affecting conservation and material condition. In addition to recovering previously unreadable variants, the work advances multispectral processing as a literary-critical act and theorises the consequences of its new methods for the study of material composition, preservation, and literary form. [1]

Extending this research from Victorian to Romantic authors, in 2025 we honed this developing technology through the test case of Percy Bysshe Shelley, analysing in particular his drowned notebook. [2] Shipwrecked with him on his final voyage, Shelley’s drowned notebook is among the most fragile items in the Bodleian Library and contains a draft of Adonais, his elegy for Keats. The significance of Shelley’s drafts is compounded for the many further poems that were never published in Shelley’s lifetime, where his notebooks remain the only witness to a poem’s existence. Selected resulting analysis and insights into Shelley’s texts have already informed the commentary of the 2025 volume of Shelley’s Complete Poetical Works.[3] The present paper releases a further tranche of results from Shelley and Christina Rossetti, while identifying the wider differences between multispectral and hyperspectral techniques for literary criticism. Through the lens of Shelley and Christina Rossetti, this paper therefore discusses the complementary roles of multispectral and hyperspectral imaging and the consequences for their digital processing in Data Cubes. It illustrates how the two methods can be fused for the purpose of literary analysis of manuscript drafts and palimpsests, with interdisciplinary implications for Digital Humanities and Textual Scholarship. As a window onto the study and survival of material texts in a digital age, the paper further proposes a model for the use of spectral imaging within textual criticism.

 

Notes

1.  Michael J Sullivan, Roger Easton and Andrew Beeby, ‘Reading Behind the Lines: Ghost Texts and Spectral Imaging in the Manuscripts of Alfred Tennyson’, The Review of English Studies, 76/324 (2025), 196–210, https://doi.org/10.1093/res/hgaf007.

2.  Michael J Sullivan & University of Oxford Public Affairs Directorate, Recovery of Literary Manuscripts, University of Oxford, https://www.youtube.com/watch?v=IvIqW-qj7eA.

3. Percy Bysshe Shelley, The Complete Poetry of Percy Bysshe Shelley, eds. Nora Crook, Neil Fraistat, Stephen Behrendt & Stuart Curran, vol. 4 (Johns Hopkins UP, 2025).

 

 

• Sarah Middle University of York,
• Émilie Pagé-Perron University of York,
• Julian Richards University of York

University of York

The ARIADNE RI is a not-for-profit archaeological Research Infrastructure founded in 2022 with more than 30 international member research institutions, including the Archaeology Data Service (UK), the Swedish National Data Service, the Institutes of Archaeology in Czechia (Brno and Prague), the Foundation for Research and Technology – Hellas, and many others. Its main service is the ARIADNE Portal, which has been developed for over a decade, with funding from the European Commission. The Portal is the main hub for discovering archaeological and related datasets provided by partners and now curates metadata for more than 4 million records. Recent enhancements to the Portal have been prompted by our involvement in European infrastructure projects, while also providing further-reaching benefits to the wider user community.

Advancing fronTier Research In the arts and hUManities (ATRIUM) is a European Union-funded project with the objective to bridge Research Infrastructures (DARIAH, ARIADNE, CLARIN, and OPERAS). As part of the research, enhancements to the ARIADNE Portal are implemented to improve the findability of archaeological and other disciplines' datasets for researchers and a wider public, in particular for citizen science practitioners. The main enhancements pertain to AO-Cat, the ontology that drives the organisation of metadata in the portal, and the Portal’s web interface itself. On the ontological front, the two main changes enabled the cataloguing of data types and the handling of new media. On the Portal side, we have seen the implementation of viewers to augment the comprehension of datasets at a glance, with a particular focus on 3D viewers (3D-Hop and Xeokit). Based on user feedback, additional features and enhancements were implemented to improve search and filtering capabilities.

Applying Reactive Twins to Enhance Monument Information Systems (ARTEMIS) is also funded by the European Union, and seeks to apply digital twins to facilitate conservation, engagement and research. A digital twin is an exact digital replica of a physical object, building or site (for example); in ARTEMIS, the Reactive Heritage Digital Twin (RHDT) takes this concept further by enabling responsive decision-making in real-time, based on sensor measurements. RHDTs are underpinned by data about the physical object, obtained through sources such as reports and scholarly publications, and modelled using the ARTEMIS ontology. Part of this work involves the use of open-source Large Language Models (LLMs) to extract relevant terminology from unstructured texts and semantically model the resulting relationships. We are additionally investigating the capability for entering natural language queries and translating them to SPARQL via an LLM, allowing for more nuanced and specialised searching. As the ARTEMIS database uses the same infrastructure as ARIADNE, these developments are planned for future integration with this more persistent resource, beyond the project’s lifespan.

In our presentation, we will start by providing some background information on ARIADNE. We will then present our case studies from ATRIUM and ARTEMIS, as well as demonstrating enhancements that are already available. To close, we will reflect on how such developments have facilitated discoverability and improved the user experience in the ARIADNE Portal, before outlining some of our future plans