Research output per year
Research output per year
-
FlashRNN: I/O-Aware Optimization of Traditional RNNs on modern hardware
Pöppel, K., Beck, M. & Hochreiter, S., 23 Apr 2025, (Accepted/In press) International Conference On Learning Representations (ICLR 2025).Research output: Chapter in Book/Report/Conference proceeding › Conference proceedings › peer-review
-
Tiled Flash Linear Attention: More Efficient Linear RNN and xLSTM Kernels
Beck, M., Pöppel, K., Lippe, P. & Hochreiter, S., 18 Mar 2025.Research output: Working paper and reports › Preprint
-
xLSTM 7B: A Recurrent LLM for Fast and Efficient Inference
Beck, M., Pöppel, K., Lippe, P., Kurle, R., Blies, P. M., Klambauer, G., Böck, S. & Hochreiter, S., 17 Mar 2025.Research output: Working paper and reports › Preprint
-
xLSTM Scaling Laws: Competitive Performance with Linear Time-Complexity
Beck, M., Schweighofer, K., Böck, S., Lehner, S. & Hochreiter, S., 02 Oct 2025, 38 p. (arXiv.org; no. 2510.02228).Research output: Working paper and reports › Preprint
Open Access -
A Large Recurrent Action Model: xLSTM enables Fast Inference for Robotics Tasks
Schmied, T., Adler, T., Patil, V., Beck, M., Pöppel, K., Brandstetter, J., Klambauer, G., Pascanu, R. & Hochreiter, S., Oct 2024, 33 p. (arXiv.org; no. 2410.22391).Research output: Working paper and reports › Preprint
Activities
- 1 Poster presentation
-
xLSTM: Extended Long Short-Term Memory
Beck, M. (Speaker), Pöppel, K. (Speaker), Spanring, M. (Speaker), Auer, A. (Speaker), Prudnikova, O. (Speaker), Kopp, M. (Speaker), Klambauer, G. (Speaker), Brandstetter, J. (Speaker) & Hochreiter, S. (Speaker)
26 Jul 2024Activity: Talk or presentation › Poster presentation › science-to-science