Prati
Benedikt Alkin
Benedikt Alkin
Potvrđena adresa e-pošte na ml.jku.at
Naslov
Citirano
Citirano
Godina
Vision-LSTM: xLSTM as Generic Vision Backbone
B Alkin, M Beck, K Pöppel, S Hochreiter, J Brandstetter
arXiv preprint arXiv:2406.04303, 2024
392024
Contrastive tuning: A little help to make masked autoencoders forget
J Lehner, B Alkin, A Fürst, E Rumetshofer, L Miklautz, S Hochreiter
Proceedings of the AAAI Conference on Artificial Intelligence 38 (4), 2965-2973, 2024
112024
Universal Physics Transformers
B Alkin, A Fürst, S Schmid, L Gruber, M Holzleitner, J Brandstetter
arXiv preprint arXiv:2402.12365, 2024
112024
Universal physics transformers: A framework for efficiently scaling neural operators
B Alkin, A Fürst, S Schmid, L Gruber, M Holzleitner, J Brandstetter
Advances in Neural Information Processing Systems 37, 25152-25194, 2025
82025
MIM-Refiner: A Contrastive Learning Boost from Intermediate Pre-Trained Representations
B Alkin, L Miklautz, S Hochreiter, J Brandstetter
arXiv preprint arXiv:2402.10093, 2024
62024
NeuralDEM-Real-time Simulation of Industrial Particulate Flows
B Alkin, T Kronlachner, S Papa, S Pirker, T Lichtenegger, J Brandstetter
arXiv preprint arXiv:2411.09678, 2024
32024
One Initialization to Rule them All: Fine-tuning via Explained Variance Adaptation
F Paischer, L Hauzenberger, T Schmied, B Alkin, MP Deisenroth, ...
arXiv preprint arXiv:2410.07170, 2024
22024
UPT++: Latent Point Set Neural Operators for Modeling System State Transitions
A Fürst, F Sestak, AP Toshev, B Alkin, NA Adams, A Mayr, G Klambauer, ...
1
Sustav trenutno ne može provesti ovu radnju. Pokušajte ponovo kasnije.
Članci 1–8