Category: Computer Science & AI
-
CORE improves link prediction by completing and reducing graph noise
This research indicates that CORE, a data augmentation method for link prediction, aims to recover missing edges and remove noisy graph structure to improve robustness and performance.
AI-assisted summary
-
Facebook news cards shaped interpretation of the border dispute
This research indicates that Facebook news card design features strongly shaped how people interpreted and reacted to the India-Bangladesh border dispute.
AI-assisted summary
-
Hybrid model improves student engagement recognition
This research indicates that a data-augmented hybrid graph convolutional network and transformer approach can recognize student engagement levels from facial activity in e-learning videos more effectively than several compared methods.
AI-assisted summary
-
AI-integrated animation teaching improved training and outcomes
This research indicates that integrating AI technology into animation teaching can improve technical application, artistic creation, and animation generation performance.
AI-assisted summary
-
Telemetry-driven fragmentation improved secure multi-cloud reconstruction resistance
This research indicates that telemetry-guided adaptive fragmentation, combined with encryption and dispersed storage, can improve predictive reliability and prevent cloud-only data reconstruction.
AI-assisted summary
-
Liger+ balances latency and throughput in distributed model inference
This research indicates that Liger+, a distributed large model inference system, can dynamically balance latency and throughput on multi-GPU architecture using interleaved parallelism.
AI-assisted summary
-
WaSC decouples WebAssembly system access with low startup latency
This research indicates that WaSC hardens WebAssembly system isolation by moving the system interface into a virtualization-based daemon while preserving low startup latency and a small memory footprint.
AI-assisted summary
-
Finnish COVID-19 monitoring data shaped equity and visibility
This research indicates that health monitoring data in Finland during COVID-19 both enabled and constrained equity-oriented responses, with disaggregated data used more fully only after extra short-term resources were allocated.
AI-assisted summary
-
DDQN improved CPU frequency control under renewable energy uncertainty
This research indicates that a Double Deep Q-Network can adapt CPU frequency scaling for edge computing under uncertain renewable energy availability while improving the balance between latency and energy use.
AI-assisted summary
-
RPA adoption requires technical, organizational, and strategic conditions
This research indicates that robotic process automation is more complex to implement than a simple, easy-to-use solution and depends on multiple technical, organizational, and strategic conditions.
AI-assisted summary