Obserwuj
Metod Jazbec
Metod Jazbec
University of Amsterdam
Zweryfikowany adres z uva.nl - Strona główna
Tytuł
Cytowane przez
Cytowane przez
Rok
Scalable gaussian process variational autoencoders
M Jazbec, M Ashman, V Fortuin, M Pearce, S Mandt, G Rätsch
International Conference on Artificial Intelligence and Statistics, 3511-3519, 2021
302021
Towards anytime classification in early-exit architectures by enforcing conditional monotonicity
M Jazbec, J Allingham, D Zhang, E Nalisnick
Advances in Neural Information Processing Systems 36, 2024
92024
Factorized Gaussian Process Variational Autoencoders
M Jazbec, M Pearce, V Fortuin
Symposium on Advances in Approximate Bayesian Inference (AABI), 2020
92020
On the impact of publicly available news and information transfer to financial markets
M Jazbec, B Pàsztor, F Faltings, N Antulov-Fantulin, PN Kolm
Royal Society Open Science 8 (7), 202321, 2021
52021
Fast yet Safe: Early-Exiting with Risk Control
M Jazbec, A Timans, TH Veljković, K Sakmann, D Zhang, CA Naesseth, ...
arXiv preprint arXiv:2405.20915, 2024
42024
Early-Exit Neural Networks with Nested Prediction Sets
M Jazbec, P Forré, S Mandt, D Zhang, E Nalisnick
The 40th Conference on Uncertainty in Artificial Intelligence, 2024
1*2024
Dynamic Vocabulary Pruning in Early-Exit LLMs
J Vincenti, KA Sadek, J Velja, M Nulli, M Jazbec
arXiv preprint arXiv:2410.18952, 2024
2024
DuoDiff: Accelerating Diffusion Models with a Dual-Backbone Approach
DG Fernández, RA Matişan, AM Muñoz, AM Vasilcoiu, J Partyka, ...
arXiv preprint arXiv:2410.09633, 2024
2024
On Efficient Distillation from LLMs to SLMs
M Jazbec, M Xia, A Mallick, D Madrigal, D Han, S Kessler, V Rühle
NeurIPS 2024 Workshop on Fine-Tuning in Modern Machine Learning: Principles …, 0
Nie można teraz wykonać tej operacji. Spróbuj ponownie później.
Prace 1–9