Marian Verhelst
Scholar

Marian Verhelst

Google Scholar ID: fa9SE3MAAAAJ
Micas - ESAT - KU Leuven, Belgium
Low-energy chip designsensor fusionmachine learningcross-layer optimization
Google Scholar↗
Citations & Impact
All-time
Citations
6,146
 
H-index
38
 
i10-index
100
 
Publications
20
 
Co-authors
43
list available
Contact
No contact links provided.
Publications
18 items
The Hyperscale Lottery: How State-Space Models Have Sacrificed Edge Efficiency
2026
Cited
0
MONET: Modeling and Optimization of neural NEtwork Training from Edge to Data Centers
2026
Cited
0
A 16 nm 1.60TOPS/W High Utilization DNN Accelerator with 3D Spatial Data Reuse and Efficient Shared Memory Access
2026
Cited
0
Torrent: A Distributed DMA for Efficient and Flexible Point-to-Multipoint Data Movement
2025
Cited
0
Hardware-aware Neural Architecture Search of Early Exiting Networks on Edge Accelerators
2025
Cited
0
The Configuration Wall: Characterization and Elimination of Accelerator Configuration Overhead
2025
Cited
0
P3-LLM: An Integrated NPU-PIM Accelerator for LLM Inference Using Hybrid Numerical Formats
2025
Cited
0
Precision-Scalable Microscaling Datapaths with Optimized Reduction Tree for Efficient NPU Integration
2025
Cited
0
Resume (English only)
Co-authors
43 total
Co-author 1
Co-author 1
Wannes Meert
Wannes Meert
DTAI, KU Leuven
Co-author 3
Co-author 3
Sofie Pollin
Sofie Pollin
Professor Electrical Engineering, KU Leuven
Co-author 5
Co-author 5
Co-author 6
Co-author 6
Co-author 7
Co-author 7
Co-author 8
Co-author 8