Category: Computer Science & AI
-
OntoLogX turns cybersecurity logs into ontology-grounded knowledge graphs
This research indicates that OntoLogX can transform raw cybersecurity logs into ontology-grounded knowledge graphs and map adversarial activity to MITRE ATT&CK tactics.
AI-assisted summary
-
New simplicial semantics for belief avoids collapsing belief into knowledge
This research indicates that a new semantics for belief in simplicial complexes can satisfy knowledge-implies-belief without making belief and knowledge identical.
AI-assisted summary
-
Embodied conversational agents improved response detail and engagement
This research indicates that embodied conversational agents can produce more informative, detailed survey responses and higher, more time-efficient engagement, while satisfaction did not change significantly.
AI-assisted summary
-
Encoding choice drives performance in hybrid quantum neural networks
This research indicates that data encoding is the main factor affecting hybrid quantum convolutional neural network performance, while measurement choices matter most in purely quantum models.
AI-assisted summary
-
Eigenvalue coalescence can be localized from loop behavior
This research indicates that generic cuspidal points, where eigenvalues of smooth complex-valued matrix functions coalesce, can sometimes be localized by examining eigenvalue periodicity and phase accumulation around loops in parameter space.
AI-assisted summary
-
Thoth improves prefetching for irregular memory access patterns
This research indicates that Thoth can uncover data-dependent memory access patterns more robustly and improve hardware prefetching on DDMA-intensive workloads.
AI-assisted summary
-
Collaborative VR authoring adds coordination overhead
This research indicates that collaborative virtual reality end-user development is technically feasible, but it is not yet seen as a suitable alternative to PC-based authoring, especially for VR-based authoring.
AI-assisted summary
-
Study maps human-centric evaluation of semantic resources
This research indicates that human-centric evaluation of semantic resources needs a theoretical framework and a systematic overview of existing approaches and trends.
AI-assisted summary
-
CORE improves link prediction with compact graph augmentations
This research indicates that CORE, a data augmentation method based on the Information Bottleneck principle, can recover missing edges and remove noise to improve link prediction.
AI-assisted summary
-
Machine learning system automates Excel chart generation
This research indicates that an ML-based dashboard can convert raw Excel spreadsheets into publication-quality graphs with faster chart selection and improved user satisfaction.
AI-assisted summary
