Research
How to Run Exam Monitoring in Italy Without Getting Fined
A practical guide to running exam monitoring in Italy, based on recent enforcement cases and the legal, transparency, data-retention, and human-review safeguards institutions need to avoid regulatory penalties.
Why UK regulators allow monitoring AI, but only with a human in the loop
An examination of the UK regulatory pathway for AI-assisted monitoring, including ICO enforcement, Ofqual guidance, human oversight, data-protection impact assessments, and accessible alternatives.
Why Monitoring AI Clears Ethics Review in Germany but Rarely Reaches Real Exams
A practical guide to running exam monitoring in Italy, based on recent enforcement cases and the legal, transparency, data-retention, and human-review safeguards institutions need to avoid regulatory penalties.
Ten Years of Research on Intelligent Educational Games for Learning Spelling and Mathematics
A retrospective on the shared architecture behind Orthograph and Calcularis, with data from more than 20,000 children.
Constructor Practice Science Labs earns ESSA Level IV evidence rating
Independent reviewers at Instructure have validated Constructor Practice Science Labs as meeting the "Demonstrates a Rationale" standard under the Every Student Succeeds Act — confirming alignment with ESSA Level IV criteria and establishing a foundation for future evaluation studies examining how Science Labs relates to student learning outcomes.
Race lens: A machine intelligence-based application for racing photo analysis
This paper presents Race lens, a novel application utilizing advanced deep learning and computer vision models for comprehensive analysis of racing photos.
SoccerNet 2022 challenges results
The SoccerNet 2022 challenges were the second annual video understanding challenges organized by the SoccerNet team.
Simultaneous perturbation method for multi-task weight optimization in one-shot meta-learning
In this paper we investigate the modification of a standard meta-learning pipeline. The proposed method simultaneously utilizes information from several meta-training tasks in a common loss function.
Efficacy of a Computer-Based Learning Program in Children with Developmental Dyscalculia. What Influences Individual Responsiveness?
This study used Calcularis 2.0 with additional training forms. It found that the subjects improved their arithmetic skills even more than with older versions of Calcularis.
Significant Percentile Improvements for Children After Calcularis Training
The above-listed publications cite the raw value and z-value improvements of the study participants, which is common for international publications.
