
TALEVA — A Platform for Evaluating Large Language Models
Building a reproducible evaluation platform for LLMs, with a particular focus on Italian

Building a reproducible evaluation platform for LLMs, with a particular focus on Italian

Interactive demo to explore the capabilites of an LLM trained on three core lexical tasks: Reverse Dictionary (generating words from definitions), Definition Modeling (producing dictionary-style definitions), and Exemplification Modeling (creating contextual usage examples)

European project building an open LLM ecosystem covering all EU languages, coordinated by ALT-EDIC

Suite of LLaMA-2 models fine-tuned for the personalized generation of counter-narratives in response to toxic comments posted on Reddit.

Data associated with the paper ‘Evaluating Large Language Models via Linguistic Profiling’