On State Reduction in Linear Attention

arXiv:2602.04852 · cs.LG · Submitted 2026-02-04 · Read on arXiv

cs.LG

Submitted: 2026-02-04

Updated: 2026-09-29

Code: https://github.com/camail-official/LinearAttentionPruning

Terminology

• Pruning — 2× in this paper · explained in 4 episodes
• Downstream Tasks — 1× in this paper · explained in 2 episodes
• Perplexity — 1× in this paper · explained in 3 episodes

Sources

• Simple linear attention language models balance the recall-throughput tradeoff
• Just read twice: closing the recall gap for recurrent language models
• Nemotron-H: A Family of Accurate and Efficient Hybrid Mamba-Transformer Models
• Quamba2: A Robust and Scalable Post-training Quantization Framework for Selective State Space Models
• Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality
• Beyond Size: How Gradients Shape Pruning Decisions in Large Language Models
• Pruner-Zero: Evolving Symbolic Pruning Metric from scratch for Large Language Models
• Hungry Hungry Hippos: Towards Language Modeling with State Space Models
• Mamba: Linear-Time Sequence Modeling with Selective State Spaces
• RULER: What's the Real Context Size of Your Long-Context Language Models?
• Language model compression with weighted low-rank factorization
• LoRA: Low-Rank Adaptation of Large Language Models
• Quantifying Memory Utilization with Effective State-Size
• Hopfield Networks is All You Need
• Retentive Network: A Successor to Transformer for Large Language Models
• Kimi Linear: An Expressive, Efficient Attention Architecture
• Test-time regression: a unifying framework for designing sequence models with associative memory
• MambaQuant: Quantizing the Mamba Family with Variance Aligned Rotation Methods
• Gated Linear Attention Transformers with Hardware-Efficient Training
• Wanda++: Pruning Large Language Models via Regional Gradients

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks