arXiv Machine Learning June 9, 2026 By Yuji Yamamoto, Satoshi Matsuura

Bit-Flip Vulnerability of Shared KV-Cache Blocks in LLM Serving Systems

Read the original on arXiv Machine Learning →

arXiv:2604. 17249v2 Announce Type: replace-cross Abstract: Rowhammer on GPU DRAM has enabled adversarial bit flips in model weights; shared KV-cache blocks in LLM serving systems present an analogous but previously unexamined target.

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 Machine Learning.

arXiv AI

Aug 11

Governing the KV Cache: Preventing Timing Side-Channel Leakage in Multi-Tenant LLM Inference

arXiv:2608. 09225v1 Announce Type: cross Abstract: The key-value (KV) cache is the primary throughput optimization in modern large language model (LLM) inference, enabling prefix reuse across requests.

By Tejasvi C. Addagada

llms efficiency benchmarks safety

arXiv AI

Sep 10

Characterizing Contention-Induced Reliability Collapse in KV-Cache Timing Side Channels for Multi-Tenant LLM Serving

arXiv:2609.06853v1 Announce Type: cross Abstract: Shared key--value (KV) cache reuse improves large language model (LLM) serving, but it can also create a timing side channel that reveals whether a p...

By Rana Abu Bakar

llms

arXiv Machine Learning

Jul 7

Walma: Learning to See Memory Corruption in WebAssembly

arXiv:2603. 24167v2 Announce Type: replace-cross Abstract: WebAssembly's (Wasm) monolithic linear memory turns a single memory-corruption bug into a bidirectional threat: a compromised module can attack its embedding host, and a malicious host can tamper with a trusted module's state.

By Oussama Draissi, Mark G\"unzel, Ahmad-Reza Sadeghi, Lucas Davi

rag safety

arXiv AI

Jul 7

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.

By Jie Li, Tongyang Wang, Yong Chen

llms efficiency

arXiv AI

Jul 23

HijackKV: New Threat in Position-Independent KV Cache Reuse

arXiv:2607. 19957v1 Announce Type: cross Abstract: Key-Value (KV) cache reduces inference latency in large language models (LLMs).

By Yichi Zhang, Zhiqi Wang, Huan Zhang, Yuchen Yang

llms efficiency

arXiv Machine Learning

Jun 19

The Correctness Illusion in LLM-Generated GPU Kernels

arXiv:2606. 20128v1 Announce Type: cross Abstract: Benchmarks for LLM-generated GPU kernels (KernelBench, TritonBench, GEAK) score correctness through fixed-shape, small-sample allclose-style checks.

By Dipankar Sarkar

llms benchmarks

One email a morning, machine-written

Related stories

Governing the KV Cache: Preventing Timing Side-Channel Leakage in Multi-Tenant LLM Inference

Characterizing Contention-Induced Reliability Collapse in KV-Cache Timing Side Channels for Multi-Tenant LLM Serving

Walma: Learning to See Memory Corruption in WebAssembly

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

HijackKV: New Threat in Position-Independent KV Cache Reuse

The Correctness Illusion in LLM-Generated GPU Kernels