Umberto Cappellazzo's picture

3 17 2

Umberto Cappellazzo

hisoka94

·

https://umbertocappellazzo.github.io/

AI & ML interests

Multimodal Large Language Models and audio-visual speech processing at @ Imperial College London.

Recent Activity

upvoted a paper 10 days ago

The Art of Scaling Reinforcement Learning Compute for LLMs

upvoted a paper 13 days ago

QeRL: Beyond Efficiency -- Quantization-enhanced Reinforcement Learning for LLMs

authored a paper 20 days ago

MoME: Mixture of Matryoshka Experts for Audio-Visual Speech Recognition

View all activity

Organizations

None yet

commented a paper 20 days ago

MoME: Mixture of Matryoshka Experts for Audio-Visual Speech Recognition

Paper • 2510.04136 • Published 22 days ago • 3 •

commented a paper 5 months ago

Scaling and Enhancing LLM-based AVSR: A Sparse Mixture of Projectors Approach

Paper • 2505.14336 • Published May 20 • 3 •

commented a paper 8 months ago

Adaptive Audio-Visual Speech Recognition via Matryoshka-Based Multimodal LLMs

Paper • 2503.06362 • Published Mar 9 • 3 •