Alessio Devoto

4 papers · 2024–2025 · 3 venues · across top CS/AI venues

Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.

Achievements

Jump to papers ↓

🌍 Conference Polyglot · Bronze 🌉 Interdisciplinary Bridge · Bronze 🗺️ Taxonomy Completionist (12) ❓ The Questioner

Venues

NAACL (2) AAAI (1) EMNLP (1)

Top co-authors

Pasquale Minervini (4) Yu Zhao (3) Aryo Pradipta Gema (2) Xiaotang Du (2) Xuanli He (2) Giwon Hong (2) Simone Scardapane (2) Emile van Krieken (1) Mohammad Reza Ghasemi Madani (1) Joshua Harris (1)

Research topics

Machine Learning (4) Artificial Intelligence (3) Core AI (3) Deep Learning (2) Application Areas (2) Optimization & Theory · Deep Learning (2) Natural Language Processing (1) Architectures (1) Learning Types (1) Models (1) Optimization & Theory · Machine Learning (1) Resources & Methods (1)

Keywords

large language model (2) transformer architecture (1) benchmark evaluation (1) attention mechanism (1) knowledge distillation (1) conditional computation (1) efficient inference (1) efficient computing (1) language understanding (1) adaptive computation (1) representation engineering (1) sparse autoencoder (1) kv cache compression (1) internal activation (1) attention score (1) knowledge conflict (1) inference-time intervention (1) transformer model (1) ground-truth annotation (1) cache optimization (1)

Papers

Adaptive Computation Modules: Granular Conditional Computation for Efficient Inference AAAI 2025 Are We Done with MMLU? NAACL 2025 Steering Knowledge Selection Behaviours in LLMs via SAE-Based Representation Engineering NAACL 2025 A Simple and Effective L_2 Norm-Based Strategy for KV Cache Compression EMNLP 2024