Zobrazeno 1 - 10
of 2 193
pro vyhledávání: '"Jazbec A."'
Publikováno v:
NeurIPS 2024 ENLSP Workshop
Increasing the size of large language models (LLMs) has been shown to lead to better performance. However, this comes at the cost of slower and more expensive inference. Early-exiting is a promising approach for improving the efficiency of LLM infere
Externí odkaz:
http://arxiv.org/abs/2410.18952
Autor:
Fernández, Daniel Gallo, Matişan, Rǎzvan-Andrei, Muñoz, Alejandro Monroy, Vasilcoiu, Ana-Maria, Partyka, Janusz, Veljković, Tin Hadži, Jazbec, Metod
Diffusion models have achieved unprecedented performance in image generation, yet they suffer from slow inference due to their iterative sampling process. To address this, early-exiting has recently been proposed, where the depth of the denoising net
Externí odkaz:
http://arxiv.org/abs/2410.09633
Autor:
Jazbec, Metod, Timans, Alexander, Veljković, Tin Hadži, Sakmann, Kaspar, Zhang, Dan, Naesseth, Christian A., Nalisnick, Eric
Publikováno v:
Advances in Neural Information Processing Systems (NeurIPS) 2024
Scaling machine learning models significantly improves their performance. However, such gains come at the cost of inference being slow and resource-intensive. Early-exit neural networks (EENNs) offer a promising solution: they accelerate inference by
Externí odkaz:
http://arxiv.org/abs/2405.20915
Early-exit neural networks (EENNs) enable adaptive and efficient inference by providing predictions at multiple stages during the forward pass. In safety-critical applications, these predictions are meaningful only when accompanied by reliable uncert
Externí odkaz:
http://arxiv.org/abs/2311.05931
Autor:
Snoj L., Ambrožič K., Čufar A., Goričanec T., Jazbec A., Lengar I., Pungerčič A., Radulović V., Rupnik S., Štancar Ž., Žerovnik G., Žohar A., Cindro V., Kramberger G., Mandić I., Mikuž M., Barbot L., Carcreff H., Destouches C., Fourmentel D., Gruel A., Villard J.F.
Publikováno v:
EPJ Web of Conferences, Vol 225, p 04031 (2020)
The JSI TRIGA reactor features several in-core and ex-core irradiation facilities, each having different properties, such as neutron/gamma flux intensity, spectra and irradiation volume. A series of experiments and calculations was performed in order
Externí odkaz:
https://doaj.org/article/7d429a5159784654a8fc86987e5cf20b
Modern predictive models are often deployed to environments in which computational budgets are dynamic. Anytime algorithms are well-suited to such environments as, at any point during computation, they can output a prediction whose quality is a funct
Externí odkaz:
http://arxiv.org/abs/2306.02652
Autor:
Saša Jazbec
Publikováno v:
Ars & Humanitas, Vol 18, Iss 1 (2024)
Umetna inteligenca je izjemno aktualna in interdisciplinarno obravnavana tema. Pričujoči prispevek obravnava denimo UI pri pouku in za pouk tujega jezika z vidika učiteljev tujega jezika. Poskusu opredelitve UI, kratkemu pregledu virov in poskusu
Externí odkaz:
https://doaj.org/article/99efba42b2ec4551a5b45d16ab853a00
Autor:
Mateja Lušnic Polak, Tomaž Polak, Mojca Kuhar, Iva Zahija Jazbec, Tadej Kaltnekar, Lea Demšar
Publikováno v:
Polish Journal of Food and Nutrition Sciences, Vol 74, Iss 2, Pp 111-119 (2024)
The effects of the supplier and the mass of pork thighs together with some production parameters including mass loss during ham production, mass of the final products and relative content of non-protein nitrogen (NPN) on the texture profile of Sloven
Externí odkaz:
https://doaj.org/article/ced88fc592b3411d948ccd7ca0390584
Publikováno v:
In International Journal of Hydrogen Energy 26 November 2024 92:560-568