RunningTensor generalizes linear attention and state‑space models by extending the recurrent memory from a second‑order tensor (matrix) to an order‑o tensor. The memory is updated via a rank‑1 outer product and read by contracting with o‑1 vector queries, with order‑2 recovering linear attention. Experiments on synthetic associative recall and real language tasks show that RunningTensor improves memory capacity from O(W²) to O(Wᵒ) and outperforms existing baselines.
By Luca Herranz-Celotti, Vincent Guigue
arXiv:2606. 01294v1 Announce Type: cross Abstract: Linear attention reduces the quadratic cost of softmax attention by maintaining a recurrent fast-weight state, but it consistently lags on in-context retrieval and long-context tasks.
By Dong Le, Thong Nguyen, Cong-Duy Nguyen, Anh Tuan Luu
arXiv:2606. 25156v1 Announce Type: new Abstract: Modern large language models based on softmax scaled-dot-product attention are constrained by their training sequence length: as the key-value sequence grows, softmax probability mass can dilute across a wider distribution, inducing activation shift and long-context performance collapse.
By Habibullah Akbar
Switching Linear Attention (SwiLA) is a new sequence layer that improves upon standard softmax attention by maintaining a fixed-size recurrent state while enhancing representational capacity. It derives its recurrence from a test-time regression framework, using online expectation-maximization in a mixture of linear regressions model. In various benchmarks—including associative recall, in-context language learning, and language modeling—SwiLA achieves strong performance, narrowing the gap to softmax attention and even surpassing it in some settings.
By Hyun Dong Lee, Xavier Gonzalez, Nicolas Zucchet, E. Kelly Buchanan, Emily B. Fox, Scott W. Linderman
arXiv:2608. 16844v1 Announce Type: cross Abstract: The quadratic cost of attention-based sequence models for long contexts has motivated a growing line of research on memory-based models that can compress context into a compact state.
By Reza Bayat, Ali Behrouz, Vahab Mirrokni, Aaron Courville
arXiv:2608. 02347v2 Announce Type: replace Abstract: Recurrent linear attention models (RLAs) such as Mamba offer efficient linear-time sequence modeling as an alternative to Transformers, yet their fixed-capacity recurrent states limit long-sequence modeling.
By Qinwen Wang, Jieping Luo, Aoxiang Qin, Ruoyu Zhao, Jianxiong Tang, Wei Zhang, Zhichao Lu, Luziwei Leng
arXiv:2606. 10650v1 Announce Type: cross Abstract: The scalability of Large Language Models (LLMs) to long contexts is fundamentally constrained by the quadratic complexity of standard attention, motivating the adoption of linear attention mechanisms with sub-quadratic cost.
By Xin Wang, Hui Shen, Boyuan Zheng, Xueshen Liu, Minkyoung Cho, Zhongwei Wan, Zesen Zhao, Zhuoqing Mao, Shen Yan, Mi Zhang
arXiv:2608. 12435v1 Announce Type: new Abstract: Transformers owe much of their strong long-context retrieval capability to a token-level memory that grows with context length.
By Ming Zhang, Kaisen Yang, Shu Yu, Ermo Hua, Ning Ding, Xia Hu, Bowen Zhou, Chaochao Lu, Youbang Sun
arXiv:2506. 05233v2 Announce Type: replace-cross Abstract: Sequence modeling is currently dominated by causal transformer architectures that use softmax self-attention.
By Johannes von Oswald, Nino Scherrer, Seijin Kobayashi, Luca Versari, Songlin Yang, Sarthak Mittal, Maximilian Schlegel, Kaitlin Maile, Yanick Schimpf, Oliver Sieberling, Alexander Meulemans, Rif A. Saurous, Guillaume Lajoie, Charlotte Frenkel, Razvan Pascanu, Blaise Ag\"uera y Arcas, Jo\~ao Sacramento
arXiv:2509. 07963v2 Announce Type: replace Abstract: The core component of attention is the scoring function, which transforms the inputs into low-dimensional queries and keys and takes the dot product of each pair.
By Yilun Kuang, Noah Amsel, Sanae Lotfi, Shikai Qiu, Andres Potapczynski, Andrew Gordon Wilson
The paper investigates how attention dynamics evolve across recurrent depth in language models, finding that attention support stabilizes early while hidden states and outputs take longer. It proposes WISE, a training‑free method that uses full attention in early steps and then reuses the discovered sparse working set for later steps, preserving performance on multi‑hop QA tasks. Experiments show that WISE maintains quality up to 2K context, offers measurable speedups, and highlights the importance of recurrent discovery of attention support.
By Ke Wan, Chen Chen
arXiv:2606. 18587v1 Announce Type: cross Abstract: Decoder-only Transformers compute attention over the KV cache of preceding tokens.
By Zhiyuan Wang, Xuan Luo, Sirui Zeng, Xifeng Yan