跳到正文
HuggingFace Daily Papers(社區熱門論文)·· 2026-10-19AI 評分37

KronLSTM:用於長上下文線性注意力的 Kronecker Delta Rule

Kronecker Delta Rule for Long Context Linear Attention

AI 導讀

KronLSTM提出 Kronecker Delta Rule,以穩定的關聯式記憶更新改善長上下文建模,無需額外正規化或記憶狀態穩定化機制。實驗顯示,它在長上下文基準測試表現尤其突出,並常勝過包括(Gated)DeltaNet 系列在內的現有線性注意力架構。其程式碼及模型已開源。

正文

Published on Oct 19

Authors:

,

,

,

,

Abstract

KronLSTM introduces a Kronecker Delta Rule for stable associative memory updates in linear attention, improving long-context modeling without extra normalization.

We propose KronLSTM, a linear attention architecture based on the Kronecker Delta Rule, a novel update rule for associative matrix memories. KronLSTM extends the xLSTM family with a memory-revision mechanism that corrects targeted associations while modifying the memory only along a single direction in matrix space. This minimally invasive memory revision reduces interference with previously stored information compared with delta-rule updates, making it particularly well suited for long-context language modeling. The resulting memory dynamics can be mathematically described as a Markov chain that converges to a steady-state distribution, while the expected norm of the memory matrix remains uniformly bounded throughout the entire input sequence. Hence, KronLSTM ensures stable memory dynamics without requiring additional normalization and stabilization mechanisms for the memory state. Experiments show that KronLSTM is competitive with, and often outperforms, existing linear attention architectures, including models of the (Gated) DeltaNet family, with particularly strong performance on long context benchmarks. These results establish the Kronecker Delta Rule as an effective and analytically tractable foundation for stable associative memory in recurrent linear attention architectures. Our code (https://github.com/NX-AI/KronLSTM) and models (https://huggingface.co/collections/NX-AI/kronlstm) are open-source.

View arXiv page View PDF Add to collection

Get this paper in your agent:

hf papers read 2610.26834

Don't have the latest CLI?

curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 4

NX-AI/KronLSTM-1.5B-100B

Text Generation •

1B •

Updated about 2 hours ago

•

14

NX-AI/KronLSTM-1.5B-100B-32k

Text Generation •

1B •

Updated about 2 hours ago

•

6

NX-AI/KronLSTM-SWA-1.5B-100B

Text Generation •

1B •

Updated about 2 hours ago

•

9

NX-AI/KronLSTM-SWA-1.5B-100B-32k

Text Generation •

1B •

Updated about 2 hours ago

•

8

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2610.26834 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2610.26834 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.

來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co