3 papers indexed
Lukas Hauzenberger, Niklas Schmidinger, Anamaria-Roberta Hartl, David Stap, Thomas Schmied, Sebastian Böck, et al.
Key-value (KV) cache growth is a major bottleneck in autoregressive decoding, as memory and bandwidth scale linearly with context length. Existing KV eviction methods often rely on static heuristics or proxy scores, which poorly track future token utility and cause brittle evicti…
Timo Bertram, Sidhant Bhavnani, Richard Freinschlag, Erich Kobler, Andreas Mayr, Günter Klambauer
In this work, we focus on SE-RRMs, a symbol-equivariant instantiation of RRMs that exhibits improved extrapolation to larger problem sizes. We propose a neuro-symbolic approach, ``Guiding with Recurrent Reasoning Models'' (G-RRM), which integrates SE-RRMs with symbolic solvers fo…
Patrick Podest, Marco Pichler, Elias Bürger, Levente Zólyomi, Bernhard Voggenberger, Wilhelm Berghammer, et al.
We introduce TiRex-2, a recurrent xLSTM-based time series foundation model that generalizes the univariate TiRex to multivariate forecasting with both past and future covariates. Real-world forecasting is inherently sequential: observations arrive continuously, variables evolve j…