2 papers
cs.CL2021
On the interaction of automatic evaluation and task framing in headline style transfer
Lorenzo De Mattei, Michele Cafagna, Huiyuan Lai +3
An ongoing debate in the NLG community concerns the best way to evaluate systems, with human evaluation often being considered the most reliable method, compared to corpus-based me…
cs.CL2020
GePpeTto Carves Italian into a Language Model
Lorenzo De Mattei, Michele Cafagna, Felice Dell'Orletta +2
In the last few years, pre-trained neural architectures have provided impressive improvements across several NLP tasks. Still, generative language models are available mainly for E…