# Gated DeltaNet-2: Decoupling Erase and Write in Linear Attention

arXiv: https://arxiv.org/abs/2605.22791 · HTML: https://mlsanity.com/paper/2605.22791/

## Key facts

- Cited by 33 papers in the map (data through 14 Oct 2026).
- Placed in Core SSM architectures for language, one of 10 directions in State-space sequence models (S4, Mamba).
- Builds on 41 of its 53 references that are in the map.
- First posted on arXiv 21 May 2026.

## Placed in

- State-space sequence models (S4, Mamba) / [Core SSM architectures for language](https://mlsanity.com/ssm/core-ssm-architectures-for-language/)

## Builds on (most cited first, 30 of 41)

- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [Mamba: Linear-Time Sequence Modeling with Selective State Spaces](https://mlsanity.com/paper/2312.00752/) (2023)
- [HellaSwag: Can a Machine Really Finish Your Sentence?](https://mlsanity.com/paper/1905.07830/) (2019)
- [Pointer Sentinel Mixture Models](https://mlsanity.com/paper/1609.07843/) (2016)
- [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)
- [Efficiently Modeling Long Sequences with Structured State Spaces](https://mlsanity.com/paper/2111.00396/) (2021)
- [BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions](https://mlsanity.com/paper/1905.10044/) (2019)
- [Can a Suit of Armor Conduct Electricity? A New Dataset for Open Book Question Answering](https://mlsanity.com/paper/1809.02789/) (2018)
- [Transformers are RNNs: Fast Autoregressive Transformers with Linear Attention](https://mlsanity.com/paper/2006.16236/) (2020)
- [Know What You Don't Know: Unanswerable Questions for SQuAD](https://mlsanity.com/paper/1806.03822/) (2018)
- [RULER: What's the Real Context Size of Your Long-Context Language Models?](https://mlsanity.com/paper/2404.06654/) (2024)
- [Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality](https://mlsanity.com/paper/2405.21060/) (2024)
- [The FineWeb Datasets: Decanting the Web for the Finest Text Data at Scale](https://mlsanity.com/paper/2406.17557/) (2024)
- [The LAMBADA dataset: Word prediction requiring a broad discourse context](https://mlsanity.com/paper/1606.06031/) (2016)
- [DROP: A Reading Comprehension Benchmark Requiring Discrete Reasoning Over Paragraphs](https://mlsanity.com/paper/1903.00161/) (2019)
- [Simplified State Space Layers for Sequence Modeling](https://mlsanity.com/paper/2208.04933/) (2022)
- [Retentive Network: A Successor to Transformer for Large Language Models](https://mlsanity.com/paper/2307.08621/) (2023)
- [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)
- [Linear Transformers Are Secretly Fast Weight Programmers](https://mlsanity.com/paper/2102.11174/) (2021)
- [Resurrecting Recurrent Neural Networks for Long Sequences](https://mlsanity.com/paper/2303.06349/) (2023)
- [Parallelizing Linear Transformers with the Delta Rule over Sequence Length](https://mlsanity.com/paper/2406.06484/) (2024)
- [Learning to (Learn at Test Time): RNNs with Expressive Hidden States](https://mlsanity.com/paper/2407.04620/) (2024)
- [Griffin: Mixing Gated Linear Recurrences with Local Attention for Efficient Language Models](https://mlsanity.com/paper/2402.19427/) (2024)
- [Repeat After Me: Transformers are Better than State Space Models at Copying](https://mlsanity.com/paper/2402.01032/) (2024)
- [Simple linear attention language models balance the recall-throughput tradeoff](https://mlsanity.com/paper/2402.18668/) (2024)
- [Zoology: Measuring and Improving Recall in Efficient Language Models](https://mlsanity.com/paper/2312.04927/) (2023)
- [Kimi Linear: An Expressive, Efficient Attention Architecture](https://mlsanity.com/paper/2510.26692/) (2025)
- [Transformer Quality in Linear Time](https://mlsanity.com/paper/2202.10447/) (2022)

## Built on by (most cited first, 30 of 33)

- [Erase-then-Delta Attention: Decoupling Erase and Write Addresses in Delta-Rule Linear Attention](https://mlsanity.com/paper/2606.26560/) (2026)
- [HydraHead: From Head-Level Functional Heterogeneity to Specialized Attention Hybridization](https://mlsanity.com/paper/2606.20097/) (2026)
- [Memory for Large Language Models](https://mlsanity.com/paper/2607.25380/) (2026)
- [Allocating Recurrent Compute in Looped Language Models](https://mlsanity.com/paper/2608.18230/) (2026)
- [AURA: Action-Gated Memory for Robot Policies at Constant VRAM](https://mlsanity.com/paper/2606.02775/) (2026)
- [MMLA: Memory-Mediated Learning Architecture for Predictive Dual-State Adaptation](https://mlsanity.com/paper/2606.28876/) (2026)
- [DART: Decoded Attention over Recurrent States for Efficient Long-Context Sequence Modeling](https://mlsanity.com/paper/2608.02032/) (2026)
- [OctoLong: Mid-Training On Cross-Repository Code Contexts Enhances Long-Context Modeling](https://mlsanity.com/paper/2608.05141/) (2026)
- [MARCH: Scaling Recurrent Memory with Content-Routed State Anchors](https://mlsanity.com/paper/2608.12435/) (2026)
- [Linear Attention Architectures: Mechanisms, Trade-offs, and Cross-Layer Routing](https://mlsanity.com/paper/2607.07953/) (2026)
- [REIMU: Efficient Heterogeneous Hierarchical Reasoning for SSL-Based Speech Deepfake Detection](https://mlsanity.com/paper/2608.00857/) (2026)
- [LiveMem: Maintaining Memory State Continuity in Long-Running LLM Inference](https://mlsanity.com/paper/2608.02515/) (2026)
- [The Query Knows What to Forget: A Second Erase Direction for Linear Attention](https://mlsanity.com/paper/2608.13668/) (2026)
- [Safin-1: Safety from Within through Memory-Native State Evolution](https://mlsanity.com/paper/2609.00092/) (2026)
- [Kalman Delta Networks: Uncertainty-aware Associative Memory](https://mlsanity.com/paper/2609.07816/) (2026)
- [Video DeltaNet: A Video-Native Hybrid Attention for Livestream Video Generation](https://mlsanity.com/paper/2609.20744/) (2026)
- [Memory by Design: Bayesian Layers for Sequence Modeling](https://mlsanity.com/paper/2605.31163/) (2026)
- [CARVE: Content-Aware Recurrent with Value Efficiency for Chunk-Parallel Linear Attention](https://mlsanity.com/paper/2606.27229/) (2026)
- [SANE: State Anomaly Neutralization for Stable Extreme-Context Delta-Rule Models](https://mlsanity.com/paper/2608.22354/) (2026)
- [TERN: A Delta-rule Memory with a Seasonal Reference and Online Adaptation for Epidemic Forecasting](https://mlsanity.com/paper/2609.18407/) (2026)
- [Low-Bit Recurrent States in Hybrid Language Models](https://mlsanity.com/paper/2609.30950/) (2026)
- [Empowering Hybrid Attention Models on NPUs](https://mlsanity.com/paper/2609.32114/) (2026)
- [CyFA: Linear Sequence Modeling with Relative-Time-Partitioned Memory](https://mlsanity.com/paper/2609.36259/) (2026)
- [Triadic Linear Attention: Three-Dimensional Recurrent States for Long-Context Sequence Modeling](https://mlsanity.com/paper/2609.36529/) (2026)
- [The Evolution of Attention in Large Language Models: Mechanisms, Trade-offs, and Emerging Trends](https://mlsanity.com/paper/2609.39661/) (2026)
- [Constant-Memory Recall: Learned Associations in a Fixed Matrix State](https://mlsanity.com/paper/2610.00232/) (2026)
- [Gated Slot Attention-2: Two-Sided Associative Memory Correction in Linear Attention](https://mlsanity.com/paper/2610.02816/) (2026)
- [Voltic: Distinguishing Volatility from Stochasticity in Recurrent Memory](https://mlsanity.com/paper/2610.05700/) (2026)
- [PHBA: Prefix-State Hybrid Block Attention](https://mlsanity.com/paper/2610.08527/) (2026)
- [Towards In-Parameter Memory Augmentation for Large Language Models](https://mlsanity.com/paper/2610.08630/) (2026)

## Abstract (arXiv)

Linear attention replaces the unbounded cache of softmax attention with a fixed-size recurrent state, reducing sequence mixing to linear time and decoding to constant memory. The hard part is not just what to forget, but how to edit this compressed memory without scrambling existing associations. Delta-rule models subtract the current read before writing a new value, and Kimi Delta Attention (KDA) sharpens forgetting with channel-wise decay. But the active edit still uses a single scalar gate to control two different things: how much old content to erase on the key side and how much new content to commit on the value side. We introduce Gated DeltaNet-2, which generalizes both Gated DeltaNet and KDA by inheriting adaptive forgetting and channel-wise decay while addressing their shared limitation, the scalar tie between erasing and writing. Gated Delta Rule-2 separates these roles with a channel-wise erase gate b_t and a channel-wise write gate w_t, reducing to KDA when both gates collapse to the same scalar and to Gated DeltaNet when the decay also collapses. We derive a fast-weight update view, a chunkwise WY algorithm with channel-wise decay absorbed into asymmetric erase factors, and a gate-aware backward pass that preserves efficient parallel training. At 1.3B parameters trained on 100B FineWeb-Edu tokens, Gated DeltaNet-2 achieves the strongest overall results among Mamba-2, Gated DeltaNet, KDA, and Mamba-3 variants across language modeling, commonsense reasoning, and retrieval. Its advantage is most pronounced on long-context RULER needle-in-a-haystack benchmarks, where it improves the evaluated multi-key retrieval setting and remains strong in both recurrent and hybrid settings. Code is available at https://github.com/NVlabs/GatedDeltaNet-2.