Karim Haroun

12 papers Journal 6Unranked 5
YearRankTypeTitle / Venue / Authors
2026 J jnl
Efficient Deep Learning for Biometrics: Overview, Challenges and Trends in Ear of Frugal AI.
CoRR
Karim Haroun, Aya Zitouni, Aicha Zenakhri, Meriem Amel Guessoum, Larbi Boubchir
2026 J jnl
Where Do Tokens Go? Understanding Pruning Behaviors in STEP at High Resolutions.
SN Comput. Sci.
Michal Szczepanski, Martyna Poreba, Karim Haroun
2025 — conf
Dynamic Hierarchical Token Merging for Vision Transformers.
VISIGRAPP (3): VISAPP
Karim Haroun, Thibault Allenet, Karim Ben Chehida, Jean Martinet
2025 — conf
Efficient Deep Learning for Biometrics: Overview, Challenges and Trends in Ear of Frugal AI.
SDS
Karim Haroun, Aya Manel Zitouni, Aicha Zenakhri, Meriem Amel Guessoum, Larbi Boubchir
2025 — conf
Exploiting Information Redundancy in Attention Maps for Extreme Quantization of Vision Transformers.
ICCVW
Lucas Maisonnave, Karim Haroun, Tom Pégeot
2025 J jnl
Exploiting Information Redundancy in Attention Maps for Extreme Quantization of Vision Transformers.
CoRR
Lucas Maisonnave, Karim Haroun, Tom Pégeot
2025 — —
Exploration de la compression dynamique dans les Vision Transformers pour la classification d'images. (Exploring dynamic compression in Vision Transformers for image classification).
Karim Haroun
2025 — conf
STEP: SuperToken and Early-Pruning for Efficient Semantic Segmentation.
VISIGRAPP (3): VISAPP
Mathilde Proust, Martyna Poreba, Michal Szczepanski, Karim Haroun
2025 J jnl
Where Do Tokens Go? Understanding Pruning Behaviors in STEP at High Resolutions.
CoRR
Michal Szczepanski, Martyna Poreba, Karim Haroun
2024 J jnl
From Near-Sensor to In-Sensor: A State-of-the-Art Review of Embedded AI Vision Systems.
Sensors
William Fabre, Karim Haroun, Vincent Lorrain, Maria Lepecq, Gilles Sicard
2021 — conf
Minimizing Subject-dependent Calibration for BCI with Riemannian Transfer Learning.
NER
Salim Khazem, Sylvain Chevallier, Quentin Barthélemy, Karim Haroun, Camille Noûs
2021 J jnl
Minimizing subject-dependent calibration for BCI with Riemannian transfer learning.
CoRR
Salim Khazem, Sylvain Chevallier, Quentin Barthélemy, Karim Haroun, Camille Noûs