# Short window attention enables long-term memorization

arXiv: https://arxiv.org/abs/2509.24552 · HTML: https://mlsanity.com/paper/2509.24552/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Builds on 38 of its 53 references that are in the map.
- First posted on arXiv 29 Sep 2025.

## Builds on (most cited first, 30 of 38)

- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Layer Normalization](https://mlsanity.com/paper/1607.06450/) (2016)
- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://mlsanity.com/paper/2104.09864/) (2021)
- [HellaSwag: Can a Machine Really Finish Your Sentence?](https://mlsanity.com/paper/1905.07830/) (2019)
- [Program Synthesis with Large Language Models](https://mlsanity.com/paper/2108.07732/) (2021)
- [Longformer: The Long-Document Transformer](https://mlsanity.com/paper/2004.05150/) (2020)
- [TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension](https://mlsanity.com/paper/1705.03551/) (2017)
- [PIQA: Reasoning about Physical Commonsense in Natural Language](https://mlsanity.com/paper/1911.11641/) (2019)
- [Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling](https://mlsanity.com/paper/1412.3555/) (2014)
- [Transformers are RNNs: Fast Autoregressive Transformers with Linear Attention](https://mlsanity.com/paper/2006.16236/) (2020)
- [LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code](https://mlsanity.com/paper/2403.07974/) (2024)
- [Root Mean Square Layer Normalization](https://mlsanity.com/paper/1910.07467/) (2019)
- [LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding](https://mlsanity.com/paper/2308.14508/) (2023)
- [Is Your Code Generated by ChatGPT Really Correct? Rigorous Evaluation of Large Language Models for Code Generation](https://mlsanity.com/paper/2305.01210/) (2023)
- [RULER: What's the Real Context Size of Your Long-Context Language Models?](https://mlsanity.com/paper/2404.06654/) (2024)
- [Sigmoid-Weighted Linear Units for Neural Network Function Approximation in Reinforcement Learning](https://mlsanity.com/paper/1702.03118/) (2017)
- [RACE: Large-scale ReAding Comprehension Dataset From Examinations](https://mlsanity.com/paper/1704.04683/) (2017)
- [Gated Delta Networks: Improving Mamba2 with Delta Rule](https://mlsanity.com/paper/2412.06464/) (2024)
- [Gated Linear Attention Transformers with Hardware-Efficient Training](https://mlsanity.com/paper/2312.06635/) (2023)
- [Hungry Hungry Hippos: Towards Language Modeling with State Space Models](https://mlsanity.com/paper/2212.14052/) (2022)
- [LongBench v2: Towards Deeper Understanding and Reasoning on Realistic Long-context Multitasks](https://mlsanity.com/paper/2412.15204/) (2024)
- [Pay Attention to MLPs](https://mlsanity.com/paper/2105.08050/) (2021)
- [Learning to (Learn at Test Time): RNNs with Expressive Hidden States](https://mlsanity.com/paper/2407.04620/) (2024)
- [Towards AI-Complete Question Answering: A Set of Prerequisite Toy Tasks](https://mlsanity.com/paper/1502.05698/) (2015)
- [BABILong: Testing the Limits of LLMs with Long Context Reasoning-in-a-Haystack](https://mlsanity.com/paper/2406.10149/) (2024)
- [Simple linear attention language models balance the recall-throughput tradeoff](https://mlsanity.com/paper/2402.18668/) (2024)
- [Samba: Simple Hybrid State Space Models for Efficient Unlimited Context Language Modeling](https://mlsanity.com/paper/2406.07522/) (2024)

## Built on by (most cited first, 11 of 11)

- [Mamba-3: Improved Sequence Modeling using State Space Principles](https://mlsanity.com/paper/2603.15569/) (2026)
- [Sparse Delta Memory: Scaling the State of Linear RNNs through Sparsity](https://mlsanity.com/paper/2607.07386/) (2026)
- [Nearly Optimal Attention Coresets](https://mlsanity.com/paper/2605.05602/) (2026)
- [Alleviating Forgetfulness of Linear Attention by Hybrid Sparse Attention and Contextualized Learnable Token Eviction](https://mlsanity.com/paper/2510.20787/) (2025)
- [Effective Distillation to Hybrid xLSTM Architectures](https://mlsanity.com/paper/2603.15590/) (2026)
- [MARCH: Scaling Recurrent Memory with Content-Routed State Anchors](https://mlsanity.com/paper/2608.12435/) (2026)
- [Sliding-window beats linear attention](https://mlsanity.com/paper/2608.28444/) (2026)
- [Safin-1: Safety from Within through Memory-Native State Evolution](https://mlsanity.com/paper/2609.00092/) (2026)
- [Do Language Models Need Sleep? Offline Recurrence for Improved Online Inference](https://mlsanity.com/paper/2605.26099/) (2026)
- [The Token Before the Value Is the Key: How Hybrid Architectures Organize Induction Circuits](https://mlsanity.com/paper/2609.15545/) (2026)
- [TraceRelay: Attention-Aligned Recurrence over Rolling Traces](https://mlsanity.com/paper/2610.11743/) (2026)

## Abstract (arXiv)

Recent works show that hybrid architectures combining local sliding window attention layers and global attention layers outperform either of these architectures taken separately. However, the impact of the window length and the interplay between local layers and global layers remain under-studied. In this work, we first analyze the interaction between short and long term memory by considering SWAX: a hybrid architecture consisting of sliding-window attention and xLSTM linear RNN layers. A counter-intuitive finding is that larger sliding windows hurts the long-context performance. In fact, short window attention encourages the model to better train the long-term memory of the xLSTM as it cannot rely on the local softmax attention mechanism for long context-retrieval. We also validate our findings on local-global architectures alternating short window and full attention layers: the short layers should be small in order not to hinder the usefulness of the long layers. However, employing too small sliding windows is detrimental even for short-context tasks, which could be solved with information from moderately larger sliding windows otherwise. Therefore, we train hybrid architectures by stochastically changing the sliding window size, forcing the model to leverage both the short term window and the long-term memory. Training with stochastic window sizes significantly outperforms regular window attention both on short and long-context problems.