Alessio Devoto

Sapienza University of Rome — Italy

Computer Science · Physical Sciences

4h-index32citations32works0.82yr avg
OpenAlex↗ORCID↗Google Scholar↗Lab page↗

Accepting Students?

No reports yet. Know if this professor is taking students?

Research Topics

Algorithms and Data Compression(4), Domain Adaptation and Few-Shot Learning(4), Neural Networks and Applications(3), Advanced Data Compression Techniques(3), Cultural Heritage Materials Analysis(3)

Publications32 total

Adaptive Semantic Token Communication for Transformer-Based Edge Inference
IEEE Transactions on Machine Learning in Communications and Networking·2026Open Access
Attention Sinks in Diffusion Language Models
arXiv (Cornell University)·2025Open Access
Universal Properties of Activation Sparsity in Modern Large Language Models
ArXiv.org·2025Open Access
Adaptive layer and token selection for efficient fine-tuning of vision transformers
Neurocomputing·2025· 2 citedOpen Access
Mixture-of-experts graph transformers for interpretable particle collision detection
Scientific Reports·2025· 2 citedOpen Access
Interpretable classification of Levantine ceramic thin sections via neural networks
Machine Learning Science and Technology·2025Open Access
Interpretable Classification of Levantine Ceramic Thin Sections via Neural Networks
arXiv (Cornell University)·2025Open Access
Class Incremental Learning with probability dampening and cascaded gated classifier
Neurocomputing·2025
Q-Filters: Leveraging Query-Key Geometry for Efficient Key-Value Cache Compression
SPIRE - Sciences Po Institutional REpository·2025Open Access
Adaptive Computation Modules: Granular Conditional Computation for Efficient Inference
Proceedings of the AAAI Conference on Artificial Intelligence·2025· 1 citedOpen Access
Q-Filters: Leveraging QK Geometry for Efficient KV Cache Compression
ArXiv.org·2025Open Access
Mixture-of-Experts Graph Transformers for Interpretable Particle Collision Detection
arXiv (Cornell University)·2025Open Access
Steering Knowledge Selection Behaviours in LLMs via SAE-Based Representation Engineering
2025Open Access
Are We Done with MMLU?
2025· 4 citedOpen Access
Exploring the Application of Interpretable Neural Networks for the Petrographic Classification of Ceramic Samples from the Levant
2025
Class Incremental Learning with Probability Dampening and Cascaded Gated Classifier
SSRN Electronic Journal·2025Open Access
Adaptive Layer and Token Selection for Efficient Fine-Tuning of Vision Transformers
SSRN Electronic Journal·2025Open Access
Goal-oriented Communications based on Recursive Early Exit Neural Networks
arXiv (Cornell University)·2024Open Access
Adaptive Semantic Token Selection for AI-native Goal-oriented Communications
2024· 4 cited
Goal-Oriented Communications Based on Recursive Early Exit Neural Networks
2024· 1 citedOpen Access
page 1 of 2Next →

Frequent Co-authors

Simone Scardapane(24), Pasquale Minervini(15), Jary Pomponi(12), Paolo Di Lorenzo(8), Aryo Pradipta Gema(5), Giwon Hong(5), Xuanli He(5), Xiaotang Du(5), Simone Petruzzi(3), Federico Alvetreti(3), Bartosz Wójcik(3), Hongru Wang(3), Kam‐Fai Wong(3), Mattia Merluzzi(3), Yu Zhao(3), Sara Capriotti(3), Silvano Mignardi(3), Laura Medeghini(3), Kamil Adamczewski(3), Joshua Ong Jun Leang(2)