firstname.lastname@upf.edu

Phone: +34 93 542 2232

Where to find me

Office 53.708
Roc Boronat building (Communication-Poblenou campus) -- when you enter the building go to the reception (left) and take the elevator to the 7th floor; my office is in front of the elevator.

With funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No 715154).

I work as a tenure-track researcher (Ramón y Cajal fellow) in the Department of Translation and Language Sciences of the Universitat Pompeu Fabra (Barcelona, Spain), and I am the head of the Computational Linguistics and Linguistic Theory (COLT) research group. 

My main research area is Computational Linguistics (aka Natural Language Processing). I am interested in semantics, or how meaning works in natural languages. I work with computational models (distributional semantics, neural networks) that are able to induce very rich and flexible semantic representations directly from examples of how people use language. I am currently working on reference, or how we use language to talk about the world: I just need to utter a bunch of words, for instance "the smart woman we met the other day at the meeting", and my interlocutor will be able to identify the person I mean. I am building computational models that are able to do this, like humans do. I am funded by an ERC Starting Grant.

Representative publications:

• Boleda, G. and A. Herbelot. 2016. Formal Distributional Semantics: Introduction to the Special Issue. Computational Linguistics 42:4, 619-635.
• McNally, L., G. Boleda. 2017. Conceptual Versus Referential Affordance in Concept Composition. In Yoad Winter & James Hampton (eds.) Compositionality and Concepts in Linguistics and Psychology, 245-267. Springer. 
• Bruni, E., G. Boleda, M. Baroni, N. K. Tran. 2012. Distributional semantics in technicolor. Proceedings of ACL 2012, pp. 136-145, Jeju Island, Korea.

For more, please check my publications, projects, resources, Google Scholar profile, and CV.

News

• March 2019: Will be an invited speaker at the RELATIONS 2019 workshop co-located with IWCS 2019, Gothenburg, Sweden, May 23.
• March 2019: Long paper, oral presentation at IWCS 2019:  Don't blame distributional semantics it can't do entailment, with Matthijs Westera.
  
• February 2019: Two papers accepted at NAACL-HLT 2019: Laura Aina, Carina Silberer, Ionut-Teodor Sorodoc, Matthijs Westera and Gemma Boleda: What do entity-centric models learn ? Insights from entity linking in multy-party dialogue; Marco Del Tredici, Raquel Fernandez and Gemma Boleda: Short-term meaning shift: a distributional exploration.
• February 2019: Will be an invited speaker at the Vector Space Semantics for Dialogue and Discourse workshop co-located with IWCS 2019, Gothenburg, Sweden, May 24.
• February 2019: Invited to participate in the Lorentz Center workshop on Compositionality in Brains and Machines, Leiden, The Netherlands, August 5-9 2019.
• December 2018: Keynote speaker at Multimodal session of Deep Learning Barcelona Symposium, Barcelona, December 20-21 2018.
• November 2018: Keynote speaker at BNAIC/BENELEARN 2018 (Benelux Conference on Artificial Intelligence and Annual Machine Learning Conference of the Benelux), ‘s-Hertogenbosch, The Netherlands, November 8-9 2018 (slides).
• October 2018: Co-organizer, Deep Learning Barcelona Symposium, Barcelona, December 20-21 2018.
• October 2018: Invited speaker at the Frontier Research and Artificial Intelligence conference organized by the ERCEA, Brussels, Belgium, 25-26 October 2018.
• August 2018: Keynote speaker at the BRIDGE and ARR workshops at ESSLLI 2018 (August 6-10, Sofia, Bulgaria). Slides here and here.
• August 2018: Short paper accepted at EMNLP 2018: How to represent a word and predict it, too: improving tied architectures for language modelling, with Kristina Gulordava and Laura Aina.