# Memory-Efficient Visual Autoregressive Modeling with Scale-Aware KV Cache Compression

arXiv: https://arxiv.org/abs/2505.19602 · HTML: https://mlsanity.com/paper/2505.19602/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- Placed in Fast autoregressive visual generation, one of 12 directions in Efficient LLM inference and serving.
- Builds on 69 of its 79 references that are in the map.
- First posted on arXiv 26 May 2025.

## Placed in

- Efficient LLM inference and serving / [Fast autoregressive visual generation](https://mlsanity.com/inference/fast-autoregressive-visual-generation/)

## Builds on (most cited first, 30 of 69)

- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis](https://mlsanity.com/paper/2307.01952/) (2023)
- [Longformer: The Long-Document Transformer](https://mlsanity.com/paper/2004.05150/) (2020)
- [Taming Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2012.09841/) (2020)
- [Efficient Streaming Language Models with Attention Sinks](https://mlsanity.com/paper/2309.17453/) (2023)
- [MaskGIT: Masked Generative Image Transformer](https://mlsanity.com/paper/2202.04200/) (2022)
- [Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction](https://mlsanity.com/paper/2404.02905/) (2024)
- [One-step Diffusion with Distribution Matching Distillation](https://mlsanity.com/paper/2311.18828/) (2023)
- [GenEval: An Object-Focused Framework for Evaluating Text-to-Image Alignment](https://mlsanity.com/paper/2310.11513/) (2023)
- [SnapKV: LLM Knows What You are Looking for Before Generation](https://mlsanity.com/paper/2404.14469/) (2024)
- [Conditional Image Generation with PixelCNN Decoders](https://mlsanity.com/paper/1606.05328/) (2016)
- [Latent Consistency Models: Synthesizing High-Resolution Images with Few-Step Inference](https://mlsanity.com/paper/2310.04378/) (2023)
- [H2O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models](https://mlsanity.com/paper/2306.14048/) (2023)
- [Emu3: Next-Token Prediction is All You Need](https://mlsanity.com/paper/2409.18869/) (2024)
- [Autoregressive Model Beats Diffusion: Llama for Scalable Image Generation](https://mlsanity.com/paper/2406.06525/) (2024)
- [Adversarial Diffusion Distillation](https://mlsanity.com/paper/2311.17042/) (2023)
- [Show-o: One Single Transformer to Unify Multimodal Understanding and Generation](https://mlsanity.com/paper/2408.12528/) (2024)
- [Autoregressive Image Generation without Vector Quantization](https://mlsanity.com/paper/2406.11838/) (2024)
- [KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache](https://mlsanity.com/paper/2402.02750/) (2024)
- [Transfusion: Predict the Next Token and Diffuse Images with One Multi-Modal Model](https://mlsanity.com/paper/2408.11039/) (2024)
- [ELLA: Equip Diffusion Models with LLM for Enhanced Semantic Alignment](https://mlsanity.com/paper/2403.05135/) (2024)
- [Model Tells You What to Discard: Adaptive KV Cache Compression for LLMs](https://mlsanity.com/paper/2310.01801/) (2023)
- [Janus: Decoupling Visual Encoding for Unified Multimodal Understanding and Generation](https://mlsanity.com/paper/2410.13848/) (2024)
- [Scissorhands: Exploiting the Persistence of Importance Hypothesis for LLM KV Cache Compression at Test Time](https://mlsanity.com/paper/2305.17118/) (2023)
- [PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling](https://mlsanity.com/paper/2406.02069/) (2024)
- [ResShift: Efficient Diffusion Model for Image Super-resolution by Residual Shifting](https://mlsanity.com/paper/2307.12348/) (2023)

## Built on by (most cited first, 22 of 22)

- [Visual Autoregressive Modeling for Instruction-Guided Image Editing](https://mlsanity.com/paper/2508.15772/) (2025)
- [Forcing-KV: Hybrid KV Cache Compression for Efficient Autoregressive Video Diffusion Models](https://mlsanity.com/paper/2605.09681/) (2026)
- [ToProVAR: Efficient Visual Autoregressive Modeling via Tri-Dimensional Entropy-Aware Semantic Analysis and Sparsity Optimization](https://mlsanity.com/paper/2602.22948/) (2026)
- [OScaR: The Occam's Razor for Extreme KV Cache Quantization in LLMs and Beyond](https://mlsanity.com/paper/2605.19660/) (2026)
- [SparVAR: Exploring Sparsity in Visual AutoRegressive Modeling for Training-Free Acceleration](https://mlsanity.com/paper/2602.04361/) (2026)
- [Multi-Scale Local Speculative Decoding for Image Generation](https://mlsanity.com/paper/2601.05149/) (2026)
- [From Static Inference to Dynamic Interaction: A Survey of Streaming Large Language Models](https://mlsanity.com/paper/2603.04592/) (2026)
- [Infinity and Beyond: Compositional Alignment in VAR and Diffusion T2I Models](https://mlsanity.com/paper/2512.11542/) (2025)
- [FasterVAR: Plug-and-Play Acceleration for Visual Autoregressive Models](https://mlsanity.com/paper/2512.16483/) (2025)
- [FastSTAR: Spatiotemporal Token Pruning for Efficient Autoregressive Video Synthesis](https://mlsanity.com/paper/2603.07192/) (2026)
- [Dynamic Mixture-of-Experts for Visual Autoregressive Model](https://mlsanity.com/paper/2510.08629/) (2025)
- [Adaptive Visual Autoregressive Acceleration via Dual-Linkage Entropy Analysis](https://mlsanity.com/paper/2602.01345/) (2026)
- [HACK++: Towards More Effective Head-Aware Key-Value Compression for Efficient Visual Autoregressive Modeling](https://mlsanity.com/paper/2606.08302/) (2026)
- [Diversity Has Always Been There in Your Visual Autoregressive Models](https://mlsanity.com/paper/2511.17074/) (2025)
- [DiverseVAR: Balancing Diversity and Quality of Next-Scale Visual Autoregressive Models](https://mlsanity.com/paper/2511.21415/) (2025)
- [StepVAR: Structure-Texture Guided Pruning for Visual Autoregressive Models](https://mlsanity.com/paper/2603.01757/) (2026)
- [Depth Adaptive Efficient Visual Autoregressive Modeling](https://mlsanity.com/paper/2604.17286/) (2026)
- [Visual Implicit Autoregressive Modeling](https://mlsanity.com/paper/2605.01220/) (2026)
- [HeatKV: Head-tuned KV-cache Compression for Visual Autoregressive Modeling](https://mlsanity.com/paper/2605.14877/) (2026)
- [Where to Refine, When to Stop: Rethinking Redundancy via Latent Discrepancy for Efficient Visual Autoregressive Generation](https://mlsanity.com/paper/2606.00310/) (2026)
- [SparSTAR: Sparse Attention for SpaceTime AutoRegressive Video Synthesis](https://mlsanity.com/paper/2608.10519/) (2026)
- [VISTA: Test-Time Compositional Alignment for Visual Autoregressive Generation](https://mlsanity.com/paper/2608.22521/) (2026)

## Abstract (arXiv)

Visual Autoregressive (VAR) modeling has garnered significant attention for its innovative next-scale prediction approach, which yields substantial improvements in efficiency, scalability, and zero-shot generalization. Nevertheless, the coarse-to-fine methodology inherent in VAR results in exponential growth of the KV cache during inference, causing considerable memory consumption and computational redundancy. To address these bottlenecks, we introduce ScaleKV, a novel KV cache compression framework tailored for VAR architectures. ScaleKV leverages two critical observations: varying cache demands across transformer layers and distinct attention patterns at different scales. Based on these insights, ScaleKV categorizes transformer layers into two functional groups: drafters and refiners. Drafters exhibit dispersed attention across multiple scales, thereby requiring greater cache capacity. Conversely, refiners focus attention on the current token map to process local details, consequently necessitating substantially reduced cache capacity. ScaleKV optimizes the multi-scale inference pipeline by identifying scale-specific drafters and refiners, facilitating differentiated cache management tailored to each scale. Evaluation on the state-of-the-art text-to-image VAR model family, Infinity, demonstrates that our approach effectively reduces the required KV cache memory to 10% while preserving pixel-level fidelity.