Zobrazeno 1 - 10
of 2 873
pro vyhledávání: '"NĚMEČEK, P."'
The indistinguishability of text generated by large language models (LLMs) from human-generated text poses significant challenges. Watermarking algorithms are potential solutions by embedding detectable signatures within LLM-generated outputs. Howeve
Externí odkaz:
http://arxiv.org/abs/2404.02138
Explainability of decisions made by AI systems is driven by both recent regulation and user demand. These decisions are often explainable only \emph{post hoc}, after the fact. In counterfactual explanations, one may ask what constitutes the best coun
Externí odkaz:
http://arxiv.org/abs/2401.14086
Tame functions are a class of nonsmooth, nonconvex functions, which feature in a wide range of applications: functions encountered in the training of deep neural networks with all common activations, value functions of mixed-integer programs, or wave
Externí odkaz:
http://arxiv.org/abs/2311.13544
In classification and forecasting with tabular data, one often utilizes tree-based models. Those can be competitive with deep neural networks on tabular data and, under some conditions, explainable. The explainability depends on the depth of the tree
Externí odkaz:
http://arxiv.org/abs/2306.06777
An important problem in reinforcement learning is designing agents that learn to solve tasks safely in an environment. A common solution is for a human expert to define either a penalty in the reward function or a cost to be minimised when reaching u
Externí odkaz:
http://arxiv.org/abs/2306.00035
Autor:
Nemecek, Adam
We argue that all building blocks of transformer models can be expressed with a single concept: combinatorial Hopf algebra. Transformer learning emerges as a result of the subtle interplay between the algebraic and coalgebraic operations of the combi
Externí odkaz:
http://arxiv.org/abs/2302.01834
Autor:
Leong, Colin, Nemecek, Joshua, Mansdorfer, Jacob, Filighera, Anna, Owodunni, Abraham, Whitenack, Daniel
Publikováno v:
EMNLP 2022
We present Bloom Library, a linguistically diverse set of multimodal and multilingual datasets for language modeling, image captioning, visual storytelling, and speech synthesis/recognition. These datasets represent either the most, or among the most
Externí odkaz:
http://arxiv.org/abs/2210.14712
Autor:
Nkadimeng, E., Mckenzie, R., Moayedi, S., Nemecek, S., Hadavand, H., Mellado, B., van Rensburg, R.
The ATLAS detector is set to undergo a substantial upgrade termed the "Phase-II" upgrade during the Long-Shutdown in preparation for the start of operation of the High Luminosity Large Hadron Collider (HL-LHC). This paper describes the development an
Externí odkaz:
http://arxiv.org/abs/2205.03374
Publikováno v:
Frontiers in Astronomy and Space Sciences, Vol 11 (2024)
Modification of the solar wind parameters at the bow shock (BS) and through the magnetosheath (MSH) is essential for the solar wind–magnetosphere interaction chain. The present study uses two approaches to determine the spatial profile of magnetic
Externí odkaz:
https://doaj.org/article/1fbbceb52f6d44da91c9aeb82c74521c
Publikováno v:
Acta Polytechnica CTU Proceedings, Vol 47 (2024)
Self-compacting concrete (SCC) is normally characterized with high pressures exerted on the formwork during casting. The pressures can easily exceed bearing capacity of a regular formwork when casting a high structural member in one step. The contrib
Externí odkaz:
https://doaj.org/article/e572e866690e4369b641bbd0ddf68dde