Technical Program - 25 June 2026

9:00 AM - 9:15 AM (BST):
Welcome and Opening Remarks

9:15 AM - 10:15 AM (BST):
Keynote 1
Session Chair: Stylianos I. Venieris (Samsung AI Center-Cambridge)

          

10:15 AM - 10:45 AM (BST):
Coffee Break

10:45 AM - 11:45 AM (BST):
Session 1 - Runtime Efficient LLM Acceleration
Session Chair: Stylianos I. Venieris (Samsung AI Center-Cambridge)

     

     
Profiling-Driven Adaptive Distributed Transformer Inference on Embedded Edge Deployment  
Muhammad Azlan Qazi (Aarhus University), Alexandros Iosifidis (Tampere University), Qi Zhang (Aarhus University)

     
SpecVocab: Speculative Decoding with a Speculative Vocabulary  
Miles Williams, Young D. Kwon, Rui Li, Alexandros Kouris, and Stylianos I. Venieris (Samsung AI Center-Cambridge)

11:45 AM - 1:00 PM (BST):
Lunch Break

1:00 PM - 2:00 PM (BST):
Keynote 2
Session Chair: Dolly Sapra (University of Amsterdam)

          

2:00 PM - 2:20 PM (BST):
Coffee Break

2:20 PM - 3:00 PM (BST):
Session 2 - Hardware-aware Efficient GenAI on Mobile and Edge Devices
Session Chair: Dolly Sapra (University of Amsterdam)

     

     

3:00 PM - 3:20 PM (BST):
Coffee Break

3:20 PM - 4:00 PM (BST)
Panel Discussion
Session Chair: Young D. Kwon (Samsung AI Center-Cambridge)

 
Themes:
     
Compress or rethink? Designing GenAI models for the edge 
     
Where should intelligence live? Revisiting the cloud-edge collaboration for GenAI 
     
From capability to utility: what does on-device GenAI genuinely enable? 
Panel Speakers:

Stefanos Laskaridis
(Amazon Science)


Hongxiang Fan
(Imperial College London)

Stylianos I. Venieris
(Samsung AI Center-Cambridge)

4:00 PM - 4:10 PM (BST):
Closing Remarks