AI & ML
When Shuffling Fails: Token Recovery from Obfuscated KV Caches
arXiv cs.CR0 views
arXiv:2611.24276v1 Announce Type: new Abstract: LLM serving systems store, offload, and share KV caches, exposing token-level representations to infrastructure operators and co-tenants. Shuffled obfuscation combines secret linear transforms with fresh permutations, claiming security from b! possible shufflings. We show that input symmetry invalidates this counting argument: security depends on distinguishable outputs under chosen inputs. Repeated tokens produce identical first-layer Value rows, making every row permutation leave the data unchanged. Only auxiliary structures such as a single-row recovery marker vary, reducing the effective state space from b! to b. We construct Repeat-to-Reveal. The attacker collects the b distinguishable states from its own probes and exploits their rank-one differences to recover an equivalent denoiser without knowing the defender's secret parameters. A one-time calibration transfers across requests sharing the same secret epoch, enabling passive token-multiset recovery from subsequently observed cache blocks through public-model dictionary matching. We instantiate the attack on official KV-Cloak and evaluate 3,000 real WildChat user prompts across nine models and eleven configurations. The attack recovers the secret parameters with high accuracy and achieves 100 percent token-multiset recovery on GPU. Refreshing row-level secrets between epochs interrupts cross-request transfer, confirming secret reuse scope as the operative vulnerability. We further confirm orbit collapse on two other open-source defenses: the sequence-permutation implementation of Thomas et al. and PermLLM. The KV-Cloak developers invited a detailed report through their private vulnerability reporting channel.
Source:https://arxiv.org/abs/2611.24276PreviousRetained Encrypted Search with Certified Split JoinsNextAutoPI: Taxonomy-Guided Adaptive Red Teaming for Prompt Injection in LLM Agents
Related posts

Bought a Mac mini to run coding agents in 2026? 10 things I'd set up on day one
0dev.to AI
Serving HTTP Accept: text/markdown, Content-Signal & /llms-full.txt on Cloudflare Pages (2026 GEO Guide)
0dev.to WebDev
CVE-2026-123890: vCenter vmdir Auth Bypass Exploited by Chinese APT Since September
0dev.to Security
None of our production checks waited for a real model reply
0dev.to DevOps
Take the five-minute drill: how fast can you reconstruct one AI approval?
0dev.to DevOps
4 Silent AWS Architecture Traps That Cause $10,000+ Bills (And the Terraform Guardrails That Stop Them)
0dev.to DevOps