arXiv AI September 3, 2026 By Renjie Xie, Juncheng Yang, Aoting Hu, Mingxi Zhang, Liyao Wu, Zheheng Hong, Wei Xu

HeadWiseKV: Budgeted Per-Head Cache Residency for Hybrid Long-Context Language Models

HeadWiseKV is a training‑free framework that compresses the residual global key–value caches of hybrid long‑context language models by assigning each physical KV head a static, multilevel history window. It formulates cache allocation as a restricted operational rate–distortion problem and uses the SeqCalib algorithm to generate per‑head residency policies that account for interactions across layers. In evaluations on four hybrid models, HeadWiseKV preserves near‑full‑KV quality while reducing peak device memory usage by 8.59% at a 112K context length and extending the largest verified context from 114K to 161K.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

Hugging Face Trending Papers

2d ago

HeadWiseKV: Budgeted Per-Head Cache Residency for Hybrid Long-Context Language Models

Long-context inference retains a growing key--value (KV) cache during decoding, which consumes substantial GPU memory and can reduce generation throughput. This bottleneck remains in hybrid language m...

llms efficiency

arXiv Machine Learning

Jun 3

Multi-Segment Attention: Enabling Efficient KV-Cache Management for Faster Large Language Model Serving

arXiv:2606. 02964v1 Announce Type: cross Abstract: Large Language Model (LLM) inference relies on key-value (KV) caches to avoid redundant attention computation.

By Chunan Shi, Yilei Chen, Yilin Chen, Xupeng Miao, Bin Cui

llms rag agents efficiency

arXiv AI

Jun 6

RedKnot: Efficient Long-Context LLM Serving with Head-Aware KV Reuse and SegPagedAttention

arXiv:2606. 06256v1 Announce Type: new Abstract: As the input length of large language model (LLM) serving continues to grow, the KV cache has become a dominant bottleneck in AI infrastructure.

By Yang Liu, ZhaoKai Luo, HuaYi Jin, ZhiYong Wang, RuoZhou He, BoYu Wang, Guanjie Chen, Junhao Hu

llms fine-tuning efficiency

Hugging Face Trending Papers

Jun 23

CompressKV: Semantic-Retrieval-Guided KV-Cache Compression for Resource-Efficient Long-Context LLM Inference

Long-context large language model (LLM) inference is increasingly constrained by the memory footprint and decoding cost of key-value (KV) caches, limiting sustainable deployment on resource-constrained hardware. Existing KV cache eviction methods typically apply heuristic token scoring over all heads in GQA-based LLMs.

llms efficiency

arXiv AI

Jun 24

CompressKV: Semantic-Retrieval-Guided KV-Cache Compression for Resource-Efficient Long-Context LLM Inference

arXiv:2606. 24467v1 Announce Type: new Abstract: Long-context large language model (LLM) inference is increasingly constrained by the memory footprint and decoding cost of key-value (KV) caches, limiting sustainable deployment on resource-constrained hardware.

By Xiaolin Lin, Jingcun Wang, Olga Kondrateva, Yiyu Shi, Bing Li, Grace Li Zhang

llms efficiency

arXiv Machine Learning

Jun 26

PersistentKV: Page-Aware Decode Scheduling for Long-Context LLM Serving on Commodity GPUs

arXiv:2606. 26666v1 Announce Type: new Abstract: Autoregressive large language model (LLM) serving is increasingly limited by key-value (KV) cache movement rather than dense matrix multiplication.

By Muhammad Ahmed

llms efficiency

One email a morning, machine-written

Related stories

HeadWiseKV: Budgeted Per-Head Cache Residency for Hybrid Long-Context Language Models

Multi-Segment Attention: Enabling Efficient KV-Cache Management for Faster Large Language Model Serving

RedKnot: Efficient Long-Context LLM Serving with Head-Aware KV Reuse and SegPagedAttention

CompressKV: Semantic-Retrieval-Guided KV-Cache Compression for Resource-Efficient Long-Context LLM Inference

CompressKV: Semantic-Retrieval-Guided KV-Cache Compression for Resource-Efficient Long-Context LLM Inference

PersistentKV: Page-Aware Decode Scheduling for Long-Context LLM Serving on Commodity GPUs