Pith. sign in

REVIEW 8 cited by

ChunkAttention: Efficient Self-Attention with Prefix-Aware KV Cache and Two-Phase Partition

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2402.15220 v4 pith:3GWDRGDS submitted 2024-02-23 cs.LG cs.CL

classification cs.LGcs.CL
keywords self-attentioncachechunkattentionmemorysystemefficientimprovekernel
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
abstract

Self-attention is an essential component of large language models (LLM) but a significant source of inference latency for long sequences. In multi-tenant LLM serving scenarios, the compute and memory operation cost of self-attention can be optimized by using the probability that multiple LLM requests have shared system prompts in prefixes. In this paper, we introduce ChunkAttention, a prefix-aware self-attention module that can detect matching prompt prefixes across multiple requests and share their key/value tensors in memory at runtime to improve the memory utilization of KV cache. This is achieved by breaking monolithic key/value tensors into smaller chunks and structuring them into the auxiliary prefix tree. Consequently, on top of the prefix-tree based KV cache, we design an efficient self-attention kernel, where a two-phase partition algorithm is implemented to improve the data locality during self-attention computation in the presence of shared system prompts. Experiments show that ChunkAttention can speed up the self-attention kernel by 3.2-4.8$\times$ compared to the state-of-the-art implementation, with the length of the system prompt ranging from 1024 to 4096.

Discussion (0). Sign in to comment.

Forward citations

Cited by 8 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Tutti: Making SSD-Backed KV Cache Practical for Long-Context LLM Serving

    cs.OS 2026-05 unverdicted novelty 7.0 of 10

    Tutti is a GPU-direct SSD-backed KV cache that removes CPU bottlenecks via object abstraction, GPU io_uring, and slack scheduling, delivering near-DRAM performance at 2x higher request rate and 27% lower cost than pri...

  2. Sparse Prefix Caching for Hybrid and Recurrent LLM Serving

    cs.LG 2026-04 unverdicted novelty 7.0 of 10

    Sparse prefix caching via dynamic programming for optimal checkpoint placement under overlap distributions improves the Pareto frontier for recurrent and hybrid LLM serving on shared-prefix data.

  3. Efficient Remote KV Cache Reuse with GPU-native Video Codec

    cs.DC 2026-02 conditional novelty 7.0 of 10

    KVCodec uses GPU-native video codecs and pipelined fetching to compress and transmit KV caches, delivering up to 3.51x faster TTFT than prior methods while preserving accuracy.

  4. MemDecay: Region-Aware KV Cache Eviction for Efficient LLM Agent Inference

    cs.LG 2026-07 conditional novelty 6.0 of 10

    Region-labeled tokens in LLM agent traces have order-of-magnitude different attention lifetimes, and a decay-plus-pinning eviction policy preserves system facts under fixed KV budgets while recency collapses as context grows.

  5. KV-RM: Regularizing KV-Cache Movement for Static-Graph LLM Serving

    cs.AR 2026-05 unverdicted novelty 6.0 of 10

    KV-RM regularizes KV-cache movement in static-graph LLM serving via block paging and merge-staged transport to improve throughput, tail latency, and memory use for variable-length decoding.

  6. SGLang: Efficient Execution of Structured Language Model Programs

    cs.AI 2023-12 conditional novelty 6.0 of 10

    SGLang is a new system that speeds up structured LLM programs by up to 6.4x using RadixAttention for KV cache reuse and compressed finite state machines for output decoding.

  7. KV-RM: Regularizing KV-Cache Movement for Static-Graph LLM Serving

    cs.AR 2026-05 unverdicted novelty 5.0 of 10

    KV-RM regularizes KV-cache movement via block paging and coalesced transfers to improve throughput, tail latency, and memory efficiency in static-graph LLM serving without changing the decoder interface.

  8. From Human Memory to AI Memory: A Survey on Memory Mechanisms in the Era of LLMs

    cs.IR 2025-04 unverdicted novelty 5.0 of 10

    The paper surveys human memory categories, maps them to LLM memory, and proposes a new three-dimension (object, form, time) categorization into eight quadrants to organize existing work and highlight open problems.

Pith tools