LCT Annual meeting 2014

The Cognitive Science Program offered by CIMeC is part of the Erasmus Mundus European Masters Program in Language and Communication Technologies ( The consortium, gathering top Universities in Computational Linguistics and related disciplines, is made of seven European Universities and two Non-European ones that together offer an integrated two-year program. The student selection is very competitive: yearly around 10 students are selected from over 300 applications arriving from all over the world.

The agenda of this year annual meeting foresees two invited talks by internationally renowned researchers, Ivan Titov (University of Amsterdam) and Fabio Massimo Zanzotto (University of Rome "Tor Vergata"), a Graduation Ceremony to celebrate the new graduates, a student conference and a seminar by Serenella Panaro (Career Coach) on how to profit of the new social networks for job seeking.

The program benefits of the strong collaboration between CIMeC, DISI and FBK, hence the consortium will meet at CIMeC (Rovereto) on the first day and at DISI and FBK (Povo) on the second day. Further information can be found on the program website.

Student Presentation Brochure

The abstract booklet of students' session could be downloaded here.


See the map.



08:30-09:00 Welcome
Room: 1st floor, Palazzo Fedrigotti
09:00-10:30 Consortium meeting
Conference Room, 1st floor, Palazzo Fedrigotti
09:00-10:30 Students private meeting
Seminar room, ground floor, Palazzo Fedrigotti
10:30-11:00 Coffee Break
Meeting room, 1st floor, Palazzo Fedrigotti
11:00-12:00 Graduation Ceremony
Conference Room, 1st floor, Palazzo Fedrigotti
12:30-14:00 Lunch Buffet
Room: 1st floor, Palazzo Fedrigotti
14:00-15:30 Invited Talk by Ivan Titov
Conference room, 1st floor, Palazzo Fedrigotti
15:30-16:00 Coffee Break
Meeting room, 1st floor, Palazzo Fedrigotti
16:00-18:00 Students Conference (posters)
20:00 Dinners
Place to be decided


09:30-11:00 Consortium meeting (partner+students)
Room: Ofek Povo 1
11:00-11:30 Coffee Break
11:30-12:30 Invited talk by Fabio Massimo Zanzotto
12:30-13:30 Lunch
13:30-14:30 Consortium Meeting (partners)
Room: Levico, Povo 2
13:30-14:30 Student visit at FBK
14:30-17:30 Seminar by Serenella Panaro (Career Coach)
Room: Ofek, Povo 1



Ivan Titov (University of Amsterdam)


Inducing Semantic Representations from Text with Little or No Supervision


Inducing meaning representations from text is one of the key objectives of NLP. Most existing statistical techniques for tackling this problem rely on large human-annotated datasets, which are expensive to create and exist only for a very limited number of languages. Even then, they are not very robust, cover only a small proportion of semantic constructions appearing in the labeled data, and are domain-dependent. We investigate Bayesian models which do not use any labeled data but induce semantic representations from unannotated texts. Unlike semantically-annotated data, unannotated texts are plentiful and available for many languages and many domains which makes our approach particularly promising. We show that these models induce linguistically-plausible semantic representations, significantly outperform current state-of-the-art approaches, and yield competitive results on question answering in the biomedical domain. We also look into several extensions of the model, and specifically consider multilingual induction of semantics, where we show that multilingual parallel texts provide a valuable source of indirect supervision for induction of shallow semantic representations, and also describe models, which induce distributed continuous representations of events.


Fabio Massimo Zanzotto (University of Rome Tor Vergata)


Distributed Tree Kernels and Distributional Semantics: between Syntactic Structures and Compositional Distributional Semantics


Syntax and word meaning contribute to the compositional construction of the meaning of sentences. In studying compositional distributional semantics, we discovered that generally models based on vectors tend to forget the syntactic structure of the sentences. Then, we started to investigate how we can partially preserve syntactic structures in vectors. We came across an interesting idea. We discovered a way to represent tree structured data in small vectors.

In this talk, I report on this idea that we called distributed tree kernels (DTK). DTKs are a novel method to reduce time and space complexity of tree kernels. Using a linear complexity algorithm to compute vectors for trees, we embed feature spaces of tree fragments in low-dimensional spaces where the kernel computation is directly done with dot product. We show that DTKs are faster, correlate with tree kernels, and obtain a statistically similar performance in two natural language processing tasks. I finally speculate on the relation between DTKs, the models on the compositional distributional semantics, and the use of distributional semantics in tree kernels.


Serenella Panaro (Career Coach)


Personal Branding Tools for your Career. Resume Writer and Trainer.


Learn about new Recruiting and Social Recruiting trends, and their impact on your job search and your career in the next years; explore the most important career tools for your Personal Branding: CVs, Resumes, LinkedIn Profiles, Cover Letters; understand how to build a Career Management Mindset being respectful of yourself and your skills; receive a personalized CV check

Info on Public Transportation From Rovereto to Povo