# GroupKV: Hierarchical KV Cache Management for Long-Context Diffusion LLM Inference

arXiv: https://arxiv.org/abs/2609.17573 · HTML: https://mlsanity.com/paper/2609.17573/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Diffusion LLM and MoE inference, one of 12 directions in Efficient LLM inference and serving.
- Builds on 33 of its 45 references that are in the map.
- First posted on arXiv 30 Jul 2026.

## Placed in

- Efficient LLM inference and serving / [Diffusion LLM and MoE inference](https://mlsanity.com/inference/diffusion-llm-and-moe-inference/)

## Builds on (most cited first, 30 of 33)

- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Efficient Streaming Language Models with Attention Sinks](https://mlsanity.com/paper/2309.17453/) (2023)
- [LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding](https://mlsanity.com/paper/2308.14508/) (2023)
- [RULER: What's the Real Context Size of Your Long-Context Language Models?](https://mlsanity.com/paper/2404.06654/) (2024)
- [Large Language Diffusion Models](https://mlsanity.com/paper/2502.09992/) (2025)
- [SnapKV: LLM Knows What You are Looking for Before Generation](https://mlsanity.com/paper/2404.14469/) (2024)
- [Simple and Effective Masked Diffusion Language Models](https://mlsanity.com/paper/2406.07524/) (2024)
- [H2O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models](https://mlsanity.com/paper/2306.14048/) (2023)
- [Discrete Diffusion Modeling by Estimating the Ratios of the Data Distribution](https://mlsanity.com/paper/2310.16834/) (2023)
- [FlexGen: High-Throughput Generative Inference of Large Language Models with a Single GPU](https://mlsanity.com/paper/2303.06865/) (2023)
- [Dream 7B: Diffusion Large Language Models](https://mlsanity.com/paper/2508.15487/) (2025)
- [Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inference](https://mlsanity.com/paper/2406.10774/) (2024)
- [Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decoding](https://mlsanity.com/paper/2505.22618/) (2025)
- [Block Diffusion: Interpolating Between Autoregressive and Diffusion Language Models](https://mlsanity.com/paper/2503.09573/) (2025)
- [DeepSpeed Inference: Enabling Efficient Inference of Transformer Models at Unprecedented Scale](https://mlsanity.com/paper/2207.00032/) (2022)
- [InfiniGen: Efficient Generative Inference of Large Language Models with Dynamic KV Cache Management](https://mlsanity.com/paper/2406.19707/) (2024)
- [Mercury: Ultra-Fast Language Models Based on Diffusion](https://mlsanity.com/paper/2506.17298/) (2025)
- [PowerInfer: Fast Large Language Model Serving with a Consumer-grade GPU](https://mlsanity.com/paper/2312.12456/) (2023)
- [dLLM-Cache: Accelerating Diffusion Large Language Models with Adaptive Caching](https://mlsanity.com/paper/2506.06295/) (2025)
- [LLaDA-V: Large Language Diffusion Models with Visual Instruction Tuning](https://mlsanity.com/paper/2505.16933/) (2025)
- [dKV-Cache: The Cache for Diffusion Language Models](https://mlsanity.com/paper/2505.15781/) (2025)
- [Reducing Transformer Key-Value Cache Size with Cross-Layer Attention](https://mlsanity.com/paper/2405.12981/) (2024)
- [RetrievalAttention: Accelerating Long-Context LLM Inference via Vector Retrieval](https://mlsanity.com/paper/2409.10516/) (2024)
- [Diffusion LLMs Can Do Faster-Than-AR Inference via Discrete Diffusion Forcing](https://mlsanity.com/paper/2508.09192/) (2025)
- [FlashDLM: Accelerating Diffusion Language Model Inference via Efficient KV Caching and Guided Diffusion](https://mlsanity.com/paper/2505.21467/) (2025)
- [Sparse-dLLM: Accelerating Diffusion LLMs with Dynamic Cache Eviction](https://mlsanity.com/paper/2508.02558/) (2025)
- [LLaDA-MoE: A Sparse MoE Diffusion Language Model](https://mlsanity.com/paper/2509.24389/) (2025)
- [FastDecode: High-Throughput GPU-Efficient LLM Serving using Heterogeneous Pipelines](https://mlsanity.com/paper/2403.11421/) (2024)
- [LongLLaDA: Unlocking Long Context Capabilities in Diffusion LLMs](https://mlsanity.com/paper/2506.14429/) (2025)
- [DPad: Efficient Diffusion Language Models with Suffix Dropout](https://mlsanity.com/paper/2508.14148/) (2025)

## Built on by (most cited first, 2 of 2)

- [d-OPD: Future-Aware On-Policy Distillation for Block Diffusion Language Models](https://mlsanity.com/paper/2609.35362/) (2026)
- [Mask-Guided KV Cache Eviction in Block Diffusion Language Models](https://mlsanity.com/paper/2610.06996/) (2026)

## Abstract (arXiv)

Diffusion large language models (dLLMs) are emerging as a promising generative paradigm that complements autoregressive decoding. In long-context settings, KV cache bloat and offloading transfer overhead have become primary bottlenecks in inference systems. Meanwhile, the periodic full-sequence recomputation and localized token updates in dLLMs make the KV lifecycle substantially more dynamic, complicating cache management and prefetch scheduling while making heavyweight token-level indexing or clustering schemes harder to amortize effectively during decoding. To address these challenges, we present textscGroupKV, a lightweight hierarchical KV cache management system for long-context dLLM inference. We observe that under block-wise decoding, tokens within the same generation block tend to access highly overlapping and spatially concentrated context regions, making group-level sparse selection effective. Building on this observation, textscGroupKV partitions the context into contiguous groups and performs coarse-to-fine sparse selection. textscGroupKV further exploits cross-layer consistency to enable predictive prefetching, and incorporates a staleness correction mechanism to maintain cache coherence under dynamic KV updates. Additionally, textscGroupKV adopts streaming prefill to reduce peak memory consumption during prefilling. Experiments show that textscGroupKV extends the maximum serviceable context length by up to 48.00× under constrained GPU memory, improves end-to-end inference performance by up to 3.73× in offload-based long-context settings, and maintains competitive task accuracy.