activity
20182026
most citedAdaptive Inference through Early-Exit Networks: Design, Challenges and Directions

106 citations · 119 across the 7 of their papers we have counts for

collaborators

12 papers

cs.RO2026

Decoupling the Declarative from the Procedural in Vision-Language-Action Models

Nikolaos Tsagkas, Andreas Sochopoulos, Chris Xiaoxuan Lu +2

Deploying generalist robotic agents in the real world requires transferable skills. Specifically, a policy trained to clone a behavior from object-specific demonstrations must gene…

cs.LG2026

WhiFlash: Accelerating Speculative Decoding with Token-Level Cross-Paradigm Routing

Young D. Kwon, Miles Williams, Rui Li +2

The autoregressive nature of large language models (LLMs) remains a significant bottleneck for inference, particularly in complex agentic workloads. While speculative decoding (SD)…

cs.CL2026

Speculative Decoding with a Speculative Vocabulary

Miles Williams, Young D. Kwon, Rui Li +2

Speculative decoding has rapidly emerged as a leading approach for accelerating language model (LM) inference, as it offers substantial speedups while yielding identical outputs. T…

cs.LG2024

Progressive Mixed-Precision Decoding for Efficient LLM Inference

Hao Mark Chen, Fuwen Tan, Alexandros Kouris +3

In spite of the great potential of large language models (LLMs) across various tasks, their deployment on resource-constrained devices remains challenging due to their excessive co…

cs.AR2022

Adaptable Butterfly Accelerator for Attention-based NNs via Hardware and Algorithm Co-design

Hongxiang Fan, Thomas Chau, Stylianos I. Venieris +5

Attention-based neural networks have become pervasive in many AI tasks. Despite their excellent algorithmic performance, the use of the attention mechanism and feed-forward network…

cs.LG2021106 cited

Adaptive Inference through Early-Exit Networks: Design, Challenges and Directions

Stefanos Laskaridis, Alexandros Kouris, Nicholas D. Lane

DNNs are becoming less and less over-parametrised due to recent advances in efficient model design, through careful hand-crafted or NAS-based methods. Relying on the fact that not…