Research activities
Translation evaluation activities
Evaluation and analysis of Large Language Models (LLMs) for the automatic translation of online encyclopaedia entries from Italian into English. Specifically, two versions of the LLM Lara recently released by Translated and one of the mechanisms on which Translated’s translation service is based, called Human in the Loop (HITL), are being evaluated.
The investigation is divided into five activities.
Research organisation: Fondazione Bruno Kessler (FBK) – Digital Industry Centre
Project leader: Dr Alessandro Cimatti
Period: February 2025 – March 2025
Mario Rossi
Introduction To verify the performance of the translation systems on content other than the Online [...]
12/05/2026
Introduction For the manual evaluation of the LaraBase and LaraFT systems, 200 sentences were rando[...]
The evaluation of the Quality Estimation (QE) functionality – i.e., the ability to automatically es[...]
In order to perform the automatic evaluation of LaraBase and LaraFT by subject class on EOL, the sa[...]
Activity 1 of the project involved the automatic evaluation of the quality of translations generate[...]
