# Enabling Dynamic Sparsity in Quantized LLM Inference

arXiv: https://arxiv.org/abs/2511.04477 · HTML: https://mlsanity.com/paper/2511.04477/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 17 references that are in the map.
- First posted on arXiv 6 Nov 2025.

## Builds on (most cited first, 12 of 12)

- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers](https://mlsanity.com/paper/2210.17323/) (2022)
- [GLU Variants Improve Transformer](https://mlsanity.com/paper/2002.05202/) (2020)
- [LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale](https://mlsanity.com/paper/2208.07339/) (2022)
- [Deja Vu: Contextual Sparsity for Efficient LLMs at Inference Time](https://mlsanity.com/paper/2310.17157/) (2023)
- [PowerInfer: Fast Large Language Model Serving with a Consumer-grade GPU](https://mlsanity.com/paper/2312.12456/) (2023)
- [AutoDroid: LLM-powered Task Automation in Android](https://mlsanity.com/paper/2308.15272/) (2023)
- [ReLU Strikes Back: Exploiting Activation Sparsity in Large Language Models](https://mlsanity.com/paper/2310.04564/) (2023)
- [Training-Free Activation Sparsity in Large Language Models](https://mlsanity.com/paper/2408.14690/) (2024)
- [R-Sparse: Rank-Aware Activation Sparsity for Efficient LLM Inference](https://mlsanity.com/paper/2504.19449/) (2025)
- [Polar Sparsity: High Throughput Batched LLM Inferencing with Scalable Contextual Sparsity](https://mlsanity.com/paper/2505.14884/) (2025)

## Built on by (most cited first, 1 of 1)

- [Activation Sparsity with Weight Approximation for Faster LLM Decoding on Offloaded Weights](https://mlsanity.com/paper/2610.02598/) (2026)

## Abstract (arXiv)

Deploying large language models (LLMs) on end-user devices is gaining importance due to benefits in responsiveness, privacy, and operational cost. Yet the limited memory and compute capability of mobile and desktop GPUs make efficient execution difficult. Recent observations suggest that the internal activations of LLMs are often dynamically sparse, meaning that for each input, only part of the network contributes significantly to the output. Such sparsity could reduce computation, but it interacts poorly with group-wise quantization, which remains the dominant approach for fitting LLMs onto resource-constrained hardware. To reconcile these two properties, this study proposes a set of techniques that realize dynamic sparse inference under low-bit quantization. The method features: (1) a zigzag-patterned quantization layout that organizes weights in a way consistent with activation sparsity and improves GPU memory locality; (2) a specialized GEMV kernel designed for this layout to fully utilize parallel compute units; and (3) a compact runtime mechanism that gathers sparse indices with minimal overhead. Across several model scales and hardware configurations, the approach achieves up to 1.55x faster decoding throughput while maintaining accuracy comparable to dense quantized inference, showing that structured sparsity and quantization can effectively coexist on commodity GPUs.