Focal Interest
Focal Interest

Benchmark (surveying)

  1. Full-resolution model reproduces key MSRE flow and neutron data
    Full-resolution multiphysics simulation of the Molten Salt Reactor Experiment couples neutronics with thermal-hydraulics to reproduce experimental behavior and reveal three-dimensional flow.
  2. AI meets Mathematics Education: Supporting Instructors in Large Mathematics Classes with Context-Aware AI
    Fine-tuned language model provides automated question answering in large calculus courses, achieving 75% accuracy. Hybrid human-AI workflows maintain instructor oversight while scaling support.
  3. AIRWAKE CONTROL OF ONR TUMBLEHOME USING CVG
    Computational study validates passive airwake control on ONR Tumblehome using columnar vortex generators to reduce turbulence and flow recirculation over flight decks.
  4. Analytic GMIB valuation was faster than Monte Carlo simulation
    Framework for valuing guaranteed minimum income benefits in variable annuities using numéraire transformation; achieves 99% computational time reduction versus Monte Carlo simulation.
  5. Detailed deduction of the Tennessee-Eastman benchmark model
    Detailed mathematical reconstruction of the Tennessee-Eastman benchmark process model with previously unavailable parameters and explicit documentation of assumptions.
  6. Enhanced grey wolf optimization improved cloud load balancing
    Enhanced optimization algorithm for cloud load balancing achieves 25% improvement in resource utilization and outperforms standard metaheuristic approaches in benchmark testing.
  7. Data augmented hybrid GCN transformer for student engagement recognition in E-learning
    Hybrid framework combining graph networks and transformers with synthetic data augmentation for automatic student engagement recognition from facial video in e-learning systems.
  8. GeoGraphNetworks provides validated spatial network data
    GeoGraphNetworks: 110 validated spatial networks spanning US and UK transportation and hydrological systems in analysis-ready JSON and XLSX formats with complete topological and geographic data.
  9. BenchPCNP provides labeled printed circuit netlist graph data
    BenchPCNP: A labeled printed circuit netlist graph dataset for partitioning benchmarking, constructed from 50 production-verified circuits with 54 distinct module labels following IPC-2612 standards.
  10. HKEN improved influential-node identification in network tests
    Algorithm for identifying influential nodes in complex networks by integrating hierarchical k-shell decomposition with extended neighborhood analysis, validated against SIR propagation models.
  11. A benchmark of expert-level academic questions to assess AI capabilities
    HLE benchmark reveals substantial gap between state-of-the-art LLMs and expert human performance on 2,500 closed-ended academic questions across mathematics, humanities, and natural sciences.