Best System Award across Tasks
The Best System Award has been established with the twofold aim of enhancing the participation of young researchers and supporting open science.
A jury composed of Giuseppe Attardi (University of Pisa), Giuseppe Castellucci (Amazon Alexa), Francesca Chiusaroli (University of Macerata), Gloria Gagliardi (University of Naples "L'Orientale") and Nicole Novielli (University of Bari) will assign the Best System Award across tasks to authors proposed by the task organizers.
Eligible systems had to have at least one young researcher (Ph.D. or undergraduate student) as a co-author, and the authors must have open-sourced their software.
Eligible systems were evaluated according to the following criteria:
- Novelty with respect to the state of the art;
- Originality (in terms of identification of new linguistic resources, identification of linguistically motivated features, and implementation of a theoretical framework grounded in linguistics);
- Critical insight, paving the way to future challenges (deep error analysis, discussion on the limits of the proposed system, discussion of the inherent challenges of the task);
- Technical soundness and methodological rigor.
The committee awarded UNITOR@SardiStance as the best system across tasks.
Paper: UNITOR@Sardistance2020: Combining Transformer-based architectures and Transfer Learning for robust Stance Detection
Authors: Simone Giorgioni, Marcello Politi, Samir Salman, Roberto Basili and Danilo Croce
Two special mentions were given to:
UmBERTo-MTSA @ AcCompl-It: Improving Complexity and Acceptability Prediction with Multi-task Learning on Self-Supervised Annotations
rmassidda @ DaDoEval: Document Dating Using Sentence Embeddings at EVALITA 2020