From 23 to 26 September 2025, Sergi Álvarez-Vidal carried out a Short-Term Scientific Mission at the Università di Salerno. His goal was both simple and ambitious: to explore how artificial intelligence could support the creation of teaching materials on neology for training in translation and terminology.

During his stay, Sergi worked closely with Mariadomenica Lo Nostro and colleagues Stefano Di Tore, Umberto Bilotti and Vincenzo Auriemma. Together, they tested the prototype of a virtual reality environment developed within ENEOLI. The experience was valuable to assess how such tools can motivate students, while also highlighting some limitations, such as usability issues and the risk of adding unnecessary complexity when applied in real classrooms.

They also experimented with agents based on large language models (LLMs) in a practical session. These were tested for their ability to generate tasks related to neology — for instance, inventing new terms or providing contextual examples — while discussing their limits, including inconsistency and the need for teacher validation. One of the conclusions was that generative models need to be paired with a structured knowledge base to ensure reliability and pedagogical soundness.

Beyond the immediate results, the stay broadened perspectives: the same methods could be adapted not only for neology but also for translator training more generally and even for creating resources to foster AI literacy. The collaboration initiated in Salerno will be key to continuing this path within ENEOLI.