[Paper] On Subquadratic Architectures: From Applications to Principles

Source: arXiv - 2606.12364v1

Overview

Transformers dominate modern sequence modeling, but their quadratic attention incurs substantial computational cost. Subquadratic architectures offer a scalable alternative. However, it remains unclear which designs yield the most effective sequence models. We compare three leading approaches: xLSTM, Mamba-2, and Gated DeltaNet. We evaluate these models on tasks with complex dependencies: (1) code-model pre-training, (2) distillation of code models from large language models, and (3) pre-training of time-series foundation models. Across these settings, xLSTM delivers the strongest overall performance. To explain xLSTM’s advantage, we present a unified formulation and analyze the underlying architectural mechanisms, focusing on state tracking and memory dynamics. Our results show that xLSTM enables more flexible and stable memory correction via its gating scheme. We corroborate these findings on controlled synthetic length-generalization tasks. Overall, our findings indicate that xLSTM’s gains on complex tasks stem from robust state tracking and accumulation.

Key Contributions

This paper presents research in the following areas:

• cs.LG

Methodology

Please refer to the full paper for detailed methodology.

Practical Implications

This research contributes to the advancement of cs.LG.

Authors

• Anamaria-Roberta Hartl
• Levente Zólyomi
• David Stap
• Pieter-Jan Hoedt
• Niklas Schmidinger
• Lukas Hauzenberger
• Sebastian Böck
• Günter Klambauer
• Sepp Hochreiter

Paper Information

• arXiv ID: 2606.12364v1
• Categories: cs.LG
• Published: June 10, 2026
• PDF: Download PDF