Understanding and Coding the KV Cache in LLMs from Scratch

Recent Developments in LLM Architectures: KV Sharing, mHC, and Compressed Attention

From Gemma 4 to DeepSeek V4, How New Open-Weight LLMs Are Reducing Long-Context Costs

From Tensor Buffer to Distributed Memory Hierarchy: A Survey of KV Cache Management for LLM Serving

arXiv:2607. 02574v1 Announce Type: cross Abstract: The key-value (KV) cache has become a first-order memory object in LLM serving rather than a temporary per-request tensor.

RKSC: Reasoning-Aware KV Cache Sharing and Confident Early Exit for Multi-Step LLM Inference

arXiv:2606. 09937v1 Announce Type: cross Abstract: We introduce RKSC (Reasoning-Aware KV Cache Sharing), a training-free inference framework that eliminates two structural redundancies in multi-branch LLM reasoning pipelines.

Mastering Long Contexts in LLMs with KVPress

RippleKV: Cross-Layer KV Cache Allocation via Perturbation Propagation

arXiv:2608. 08684v1 Announce Type: cross Abstract: Long-context LLM inference is bottlenecked by KV cache memory, yet distributing a limited cache budget across layers remains challenging.

PTStore (Prefix Tensor Store): Distributed Prefix Caching and Replication for High Throughput Inference Serving

arXiv:2607. 22648v1 Announce Type: new Abstract: Inspired by the design of client caching in Content Delivery Networks (CDNs), PTStore distributes and replicates popular tensors that form reusable KV cache prefixes, which are the main technique used by state of art approaches to accelerate inferences.

Cross-Model KV Cache Transfer in LLM Families: A Closed-Form Linear Mapping for Prefill Reuse

arXiv:2608. 03893v1 Announce Type: new Abstract: Production deployments often swap between different-sized models in a family for cost-quality cascading, mid-conversation switching, and routing, and each swap forces the receiver to repay the prefill from scratch.

FreqDepthKV: Frequency-Guided Depth Sharing for Robust KV Cache Compression in Long-Context LLM Inference

arXiv:2607. 06519v1 Announce Type: new Abstract: Long-context LLM inference is increasingly limited by the memory and bandwidth cost of KV caches, yet aggressive compression can remove the layer-specific evidence needed for retrieval and multi-step reasoning.

Back from the Future: Key-Value Cache Management by Counter-Causal Surprise

arXiv:2607. 27600v1 Announce Type: new Abstract: Key-value (KV) cache management through compression and eviction strategies has emerged as an important research direction in recent years.

KV Cache from scratch in nanoVLM

ReFreeKV: Towards Threshold-Free KV Cache Compression

arXiv:2502. 16886v4 Announce Type: replace-cross Abstract: To reduce memory consumption during LLM inference, a handful of methods have been proposed for KV cache pruning.

Sub-Token Routing for KV Cache Compression

arXiv:2604. 21335v3 Announce Type: replace Abstract: Transformer inference often requires a large KV cache, especially for long-context language modeling and multimodal generation.