Language and Vision are two fundamental modalities through which human beings acquire knowledge about the world. We see and speak about things and events around us, and by doing so, we learn properties and relations about objects. These two modalities are quite interdipendent and we constantly mix information we acquire through them. However, computational models of language and vision have been developing separately and the two research comunities have for a long time been unaware of each other's work. Interestingly, through these parallel research lines, they have developed highly compatible representations of words and images, respectively.
The importance of developing computational models of language and vision together has been highligted by philosophers and cognitive scientists since the birth of the Aritificial Intelligence paradigm. Only recently, however, the challenge has been empirically taken up by computational linguists and computer vision researchers.
In the last two decades, the availability of large amounts of text on the web has led to tremendous improvements in NLP research. Sophisticated textual search engines are now well consolidated and part of everybody's daily life. Images are the natural next challenge of the digital society. The combination of language and vision is the winning horse for this new era.
The UniTrento researchers are at the fronting edges of this new challenge. Driven by theoretical questions, we look at applications as the test bed for our models. The focus so far has been on the investigation of multimodal models combining linguistic and visual vector representations; cross-modal mapping from visual to language space; enhancement of visual recognizers through language models. The recent results on LaVI at UniTrento have profitted on the close collaboartion with the team of the ERC project COMPOSES and of MHUG Research Group.
We are part of the Cost Action The European Network on Integrating Vision and Language
- Raffaella Bernardi (Language-vision area coordinator, senior researcher)
- Marco Baroni (senior researcher, on leave -- at Facebook)
- Aurelie Herbelot (senior researcher)
- Sandro Pezzelle (PhD student)
- Ravi Shekhar (PhD student)
- Ionut-Teodor Sorodoc, research assistant
- Yauhen Klimovich (EM LCT student)
- Gemma Boleda (senior researcher)
- Angeliki Lazaridou (former PhD student)
- Elia Bruni (former PhD)
- Le Diu Thu (former PhD)
- Giovanni Cassani (former MSc student)
- Dat Tien Nguyen (EM LCT student)
- Laura Bostan (EM LCT student)
- Ionut Teodor Sorodoc, Sandro Pezzelle, Mariella Dimiccoli, Aurelie Herbelot and Raffaella Bernardi Pay attention to those sets! Learning quantification from images Submitted.
- Ravi Shanker, Sandro Pezzelle, Yauhen Klimovich, Aurélie Herbelot, Moin Nabi, Enver Sangineto and Raffaella Bernardi FOIL it! Find One mismatch between Image and Language caption ACL 2017
- Sandro Pezzelle, Marco Marelli and Raffaella Bernardi Be Precise or Fuzzy: Learning the Meaning of Cardinals and Quantifiers from Vision. Short paper at EACL 2017
- Angeliki Lazaridou, Marco Marelli and Marco Baroni Multimodal word meaning induction from minimal exposure to natural text. Cognitive Science, to appear.
- Sorodoc, I., Lazaridou, A., Boleda, G., Herbelot, A., Pezzelle, S. and Bernardi, R. 2016. "Look, some green circles!": Learning to quantify from images. In proceedings of the 5th Workshop on Vision and Language (collocated with ACL2016), Berlin, Germany.
- Sandro Pezzelle, Ravi Shekhar and Raffaella Bernardi Building a bagpipe with a bag and a pipe: Exploring Conceptual Combination in Vision In Proceedings of the 5th Workshop on Vision and Language, pages 60–64, Berlin, Germany, August 12 2016. Association for Computational Linguistics.
- Angeliki Lazaridou, Nghia The Pham and Marco Baroni Towards Multi-Agent Communication-Based Language Learning
- Angeliki Lazaridou, Nghia The Pham and Marco Baroni The red one!: On learning to refer to things based on their discriminative properties. ACL 2016 Short, Oral
- Boleda, G., S. Padó and M. Baroni. 2016. "Show me the cup": Reference with with continuous representations. arXiv e-print 1606.08777.
- Angeliki Lazaridou, Grzegorz Chrupala, Raquel Fernandez and Marco Baroni Multimodal semantic learning from child-directed input NAACL 2016 Short
- Marco Baroni (Submitted) Enhancing Distributional Semantics with Visual.
- Angeliki Lazaridou, Georgiana Dinu, Adam Liska and Marco Baroni, (TACL 2015) From visual attributes to adjectives through decompositional distributional semantics
- Angeliki Lazaridou, Nghia The Pham and Marco Baroni (NAACL 2015) Combining Language and Vision with a Multimodal Skip-gram Model
- Dieu-Thu Le, Jasper Uijlings and Raffaella Bernardi (Submitted). Extracting knowledge from text corpora to aid visual recognition in images.
- E. Bruni, M. Baroni, G.B. Tran Multimodal Distributional Semantics. In Journal of Artificial Intelligence Research, vol. 49, pp. 1-17, 2014.
- Angeliki Lazaridou, Elia Bruni and Marco Baroni. (2014) Is this a wampimuk? Cross-modal mapping between distributional semantics and the visual world. Proceedings of ACL 2014 (52nd Annual Meeting of the Association for Computational Linguistics), East Stroudsburg PA: ACL, 1403-1414.
- Dat T. Nguyen, Angeliki Lazaridou and Raffaella Bernardi. (2014) Coloring objects: Adjective-noun visual semantic compositionality. Proceedings of VL (Third Workshop on Vision and Language), East Stroudsburg PA: ACL, 112-114.
- Dieu-Thu Le, Jasper Uijlings and Raffaella Bernardi (2014) TUHOI: Trento Universal Human Object Interaction Dataset Proceedings of the Third Workshop on Vision and Language. Dublin City University and the Association for Computational Linguistics. pp. 17-24, Dublin [Dataset]
- A. Anderson, E. Bruni, U. Bordignon, M. Poesio and M. Baroni. (2013) Of words, eyes and brains: Correlating image-based distributional semantic models with neural representations of concepts. Proceedings of EMNLP 2013 (Conference on Empirical Methods in Natural Language Processing), East Stroudsburg PA: ACL, 1960-1970.
- Dieu-Thu Le, Jasper Uijlings and Raffaella Bernardi (2013) Exploiting language models for visual recognition EMNLP 2013
- Dieu-Thu Le, Raffaella Bernardi and Jasper Uijlings (2013) Exploiting language models to recognize unseen actions, ACM International Conference on Multimedia Retrieval, Dallas, Texas, USA, 2013.
- E. Bruni, J. Uijlings, M. Baroni and N. Sebe. (2012) Distributional semantics with eyes: Using image analysis to improve computational representations of word meaning. Brave New Idea paper. Proceedings of MM 12 (20th ACM International Conference on Multimedia), New York NY: ACM, 1219-1228.
- E. Bruni, G. Boleda, M. Baroni and N. Tran (2012) Distributional semantics in technicolor. Proceedings of ACL 2012 (50th Annual Meeting of the Association for Computational Linguistics), East Stroudsburg PA: ACL, 136-145. The data sets from this study.
- E. Bruni, G.B. Tran and M. Baroni (2011) Distributional semantics from text and images. Proceedings of the EMNLP 2011 Geometrical Models for Natural Language Semantics (GEMS). Workshop, East Stroudsburg PA: ACL, 22-32.