Ludovica Pannitto
Scroll
About me
I am the NLP Lab Manager at Laboratorio Sperimentale, a laboratory at LILEC, University of Bologna devoted to experimental, computational, and applied research in linguistics, translation, and cultural studies.
I hold a PhD in Cognitive and Brain Sciences from CIMeC, University of Trento, where I worked under the supervision of Aurélie Herbelot.
My research lies at the intersection of linguistics and natural language processing, with a focus on the grammar of spoken language, computational construction grammar, and cognitive linguistics. I am particularly interested in studying language models from a linguistically informed, usage-based perspective.
Currently I’m also Adjunct Professor at University of Salerno, teaching “Basic Computer Science” (Informatica di Base, 9 CFU) and at University of Pisa, co-teaching “Fondamenti Teorici e Programmazione” for the degree in Informatica Umanistica, together with Barbara Leporini and Salvatore Citraro.
Alongside research, I am committed to teaching, community work, and science communication.
I love vintage computer science.
Everything in life is a work in progress, this place included.
Education
PhD in Cognitive and Brain Sciences
CIMeC, University of Trento
Rovereto
November 2018 to May 2023
During my PhD, I focused on developing a paradigm to evaluate neural language models from a constructionist perspective.
Dissertation: CALaMo: a Constructionist Perspective on the Analysis of Linguistic Behaviour of Language Models
MSc in Digital Humanities - Language Technologies
University of Pisa
Pisa
July 2016 to October 2018
Dissertation: Event Knowledge in Compositional Distributional Semantics
BA in Digital Humanities
University of Pisa
Pisa
October 2013 to July 2016
Dissertation: LISA: a Light Italian Semantic Analyzer
Advanced Training
- FieldLing - International School in Linguistics Fieldwork — Sep 2026, INALCO, Paris, France
- 2nd UniDive training school on Linguistic Diversity in NLP — Jan 2026, Yerevan, Armenia
- Julia High Performance — Feb 2025, CINECA, Bologna, Italy
- Introduction to Fortran for Scientific Computing — Nov 2024, CINECA, Bologna, Italy
- 1st UniDive training school — Jul 2024, Chișinău, Moldova
- ASPP — 14th Advanced Scientific Programming in Python — Aug 2022, Bilbao, Spain
- ESSLLI — 30th European Summer School in Logic, Language and Information — Aug 2018, Sofia, Bulgaria
Research Interests
- Spoken language and interaction: the grammar of spoken Italian and conversational syntax, from co-constructions to interactional phenomena, mostly through the KIParla corpus.
- Corpus design and annotation: Universal Dependencies for spoken data, multimodal and gesture data, annotation tools, and reusable corpus architectures.
- Speech technology for conversation: what automatic speech recognition gets right and wrong on spontaneous talk, and how semi-automatic transcription can support corpus creation.
- Computational construction grammar: modelling and annotating constructions (the Italian Constructicon, NPN patterns) and probing how language models represent them.
- Language models through a linguistic lens: constructionist and usage-based evaluation of what models learn, including grammar acquisition from limited input.
- Distributional semantics and usage-based linguistics: how meaning and grammar emerge from patterns of use, from distributional models of word and sentence meaning (event knowledge, thematic fit, hypernymy, adjective classes) to usage-based accounts of grammar.
Supervision
- Il parlato spontaneo nell’interlingua di un apprendente L1 bangla: un’analisi corpus-based per la didattica dell’italiano L2 — Sara D’Ascenzo Master thesis, 2026, co-supervision with Prof. Rosa Pugliese
- ‘Layer su Layer’: identifying and disambiguating the Italian NPN construction through BERT’s contextual embeddings — Greta Gorzoni Master thesis, 2025, co-supervision with Prof. Francesca Masini
- The Fate of Interaction after ASR Decoding Optimization — Martina Simonotti Master thesis, 2025, co-supervision with Prof. Adriano Ferraresi, Prof. Caterina Mauri, Prof. Maja Miličević Petrović
- Categorizzazione ad-hoc e gestualità: una indagine corpus-based sul corpus gest-it — Virginia Vento Master thesis, 2025, co-supervision with Prof. Caterina Mauri
Community
Organising
- International conference “Talking Data”: Methodological and theoretical challenges raised by spoken interaction data — Organizing Committee, 2025, Bologna, Italy
- CLiC-it: Eleventh Italian Conference on Computational Linguistics — Senior Program Committee, 2025, Cagliari, Italy
- SpLAn-UD: UniDive Workshop on Spoken Language Annotation for Universal Dependencies — Organizing Committee, 2025, Bologna, Italy
- Doctoral Student Day — Organizing Committee, 2021, online
Reviewing
I serve as reviewer for: ACL-IJCNLP, COLING, EACL, EMNLP, IWCS, LREC-COLING, CMCL, CLiC-it, LiNCR, DANKMEMES, PRSLLS, PACLIC, RANLP, GWC, Constructions and Frames.
Networks and projects
- UniDive COST Action CA21167: Active member contributing to Working Group 1 (WG1) and Working Group 2 (WG2). Co-lead of Task 1.5: Annotation of Spoken Data (with Sylvain Kahane and Kaja Dobrovoljc) and Task 2.5: Extensions of UD-PARSEME annotation for constructions (with Francesca Masini).
- KIParla Project: Data Manager for the KIParla corpus, overseeing computational developments and annotation strategies for the first UD treebank of spoken Italian.
- Italian Association for Computational Linguistics (AILC): Active collaborator in dissemination, public engagement, and development of activities aimed at increasing the visibility of computational linguistics in Italy.
Toolbox
Languages: Italian (native), English (fluent, C2), Italian Sign Language – LIS (fluent), French (B1), German (B1)
Programming and Technical Skills: - Programming languages: Python (advanced), Bash (advanced), Regular Expressions (advanced), Julia (basic), Fortran (basic) - Data management: Git (advanced), Linux distributions (advanced), corpus creation and annotation, reproducible workflows (GitOps) - Natural Language Processing: Distributional Semantics, corpus processing, syntax and semantic annotation, computational modeling of language - Tools and technologies: NLTK, spaCy, Hugging Face Transformers, SketchEngine, Universal Dependency tools, TEI encoding tools
Soft Skills: Project management and leadership in interdisciplinary research settings; mentoring and supervision of students in linguistics, computational linguistics, and NLP; public speaking, science communication, and outreach; collaborative work in international, cross-disciplinary teams.