Web pages of R. Bernardi at UniTN
Freely available Resources
- BoB (Bolzano Bot) Dialogues
- The SICK data set for large-scale evaluation of compositional semantic models
- Dataset on alternatives See G. Kruszewski, D. Paperno, R. Bernardi and M. Baroni 2016 at the Composes website.
Language and Vision
- 89 Pascal Action Dataset (20 objects) images annotated w.r.t. actions and scenes
- TUHOI (Trento Universal Human Object Interaction Dataset)
- QUANTities in Images and Text
- FOIL it: Find one Miss-match between caption and image
- VISTA: Visually-grounded Talking Agents
- CONTINUAL-VISTA: CONTINUAL Learning for VISually-grounded Talking Agents
- Grounded Textual Entailment.