User profiles for Silvia Terragni

Silvia Terragni

Upwork
Verified email at upwork.com
Cited by 918

Cross-lingual contextualized topic models with zero-shot learning

F Bianchi, S Terragni, D Hovy, D Nozza… - arXiv preprint arXiv …, 2020 - arxiv.org
Many data sets (eg, reviews, forums, news, etc.) exist parallelly in multiple languages. They
all cover the same content, but the linguistic differences make it impossible to use traditional, …

Pre-training is a hot topic: Contextualized document embeddings improve topic coherence

F Bianchi, S Terragni, D Hovy - arXiv preprint arXiv:2004.03974, 2020 - arxiv.org
Topic models extract groups of words from documents, whose interpretation as a topic
hopefully allows for a better understanding of the data. However, the resulting word groups are …

In-context learning user simulators for task-oriented dialog systems

S Terragni, M Filipavicius, N Khau, B Guedes… - arXiv preprint arXiv …, 2023 - arxiv.org
This paper presents a novel application of large language models in user simulation for task-oriented
dialog systems, specifically focusing on an in-context learning approach. By …

Contrastive language-image pre-training for the italian language

F Bianchi, G Attanasio, R Pisoni, S Terragni… - arXiv preprint arXiv …, 2021 - arxiv.org
CLIP (Contrastive Language-Image Pre-training) is a very recent multi-modal model that jointly
learns representations of images and texts. The model is trained on a massive amount of …

Which matters most? comparing the impact of concept and document relationships in topic models

S Terragni, D Nozza, E Fersini… - Proceedings of the First …, 2020 - aclanthology.org
Topic models have been widely used to discover hidden topics in a collection of documents.
In this paper, we propose to investigate the role of two different types of relational …

Constrained relational topic models

S Terragni, E Fersini, E Messina - Information Sciences, 2020 - Elsevier
Relational topic models (RTM) have been widely used to discover hidden topics in a collection
of networked documents. In this paper, we introduce the class of Constrained Relational …

OCTIS: Comparing and optimizing topic models is simple!

S Terragni, E Fersini, BG Galuzzi… - Proceedings of the …, 2021 - aclanthology.org
In this paper, we present OCTIS, a framework for training, analyzing, and comparing Topic
Models, whose optimal hyper-parameters are estimated using a Bayesian Optimization …

Contrastive language and vision learning of general fashion concepts

PJ Chia, G Attanasio, F Bianchi, S Terragni… - Scientific Reports, 2022 - nature.com
The steady rise of online shopping goes hand in hand with the development of increasingly
complex ML and NLP models. While most use cases are cast as specialized supervised …

BETOLD: A task-oriented dialog dataset for breakdown detection

S Terragni, B Guedes, A Manso… - Proceedings of the …, 2022 - aclanthology.org
Task-Oriented Dialog (TOD) systems often suffer from dialog breakdowns-situations in which
users cannot or do not want to proceed with the conversation. Ideally TOD systems should …

Word embedding-based topic similarity measures

S Terragni, E Fersini, E Messina - … on applications of Natural Language to …, 2021 - Springer
Topic models aim at discovering a set of hidden themes in a text corpus. A user might be
interested in identifying the most similar topics of a given theme of interest. To accomplish this …