no code implementations • EACL (BEA) 2021 • Luca Benedetto, Giovanni Aradelli, Paolo Cremonesi, Andrea Cappelli, Andrea Giussani, Roberto Turrin
Classical approaches to question calibration are either subjective or require newly created questions to be deployed before being calibrated.
no code implementations • RANLP 2021 • Ekaterina Loginova, Luca Benedetto, Dries Benoit, Paolo Cremonesi
They use questions of known difficulty to train models capable of inferring the difficulty of questions from their text.
no code implementations • 17 Jul 2023 • Andrew Caines, Luca Benedetto, Shiva Taslimipoor, Christopher Davis, Yuan Gao, Oeistein Andersen, Zheng Yuan, Mark Elliott, Russell Moore, Christopher Bryant, Marek Rei, Helen Yannakoudakis, Andrew Mullooly, Diane Nicholls, Paula Buttery
The recent release of very large language models such as PaLM and GPT-4 has made an unprecedented impact in the popular media and public consciousness, giving rise to a mixture of excitement and fear as to their capabilities and potential uses, and shining a light on natural language processing research which had not previously received so much attention.
1 code implementation • 17 May 2023 • Luca Benedetto
We find that Transformer based models are the best performing across different educational domains, with DistilBERT performing almost as well as BERT, and that they outperform other approaches even on smaller datasets.
no code implementations • 28 Jun 2021 • Luca Benedetto, Paolo Fantozzi, Luigi Laura
Then, we leverage the errors of such model and train a second Transformer-based model to partition the problem instances into groups of different complexity, thus detecting the ones that can be solved without using too expensive resources.
1 code implementation • 28 Apr 2020 • Luca Benedetto, Andrea Cappelli, Roberto Turrin, Paolo Cremonesi
Statistical models such as those derived from Item Response Theory (IRT) enable the assessment of students on a specific subject, which can be useful for several purposes (e. g., learning path customization, drop-out prediction).
1 code implementation • 21 Jan 2020 • Luca Benedetto, Andrea Cappelli, Roberto Turrin, Paolo Cremonesi
The main objective of exams consists in performing an assessment of students' expertise on a specific subject.