It is shown that input DP is orthogonal to model confidentiality and that the local-DP premise required for shuffle amplification cannot hold under correctness-bounded noise, and that the leaked spectra enable fingerprinting, lineage attribution, and improved logit-based extraction, while suppressing them destroys inference utility.
Abstract
Hybrid fully homomorphic encryption~(FHE) inference improves the practicality of private inference by letting the server evaluate linear layers homomorphically while the client decrypts and applies nonlinearities. Recent schemes attempt to protect model confidentiality by returning noisy, output-permuted responses and appealing to shuffle-model differential privacy~(DP). We show that this protection fails in the correctness regime required by hybrid FHE systems. For a $d$-input linear layer, $d+1$ admissible queries suffice for exact recovery of a permutation-invariant layer summary, hence for perfect model distinguishability. We further show that input DP is orthogonal to model confidentiality and that the local-DP premise required for shuffle amplification cannot hold under correctness-bounded noise. We recover all linear layers of a \safhire{}-style ResNet-20 end-to-end from TFHE transcripts with zero error, using $d+1$ queries per layer for a total of $5{,}712$ direct queries. Under the same query model, we also confirm exact per-layer recovery on pretrained ImageNet-scale CNNs and ViT-B/16. The leaked spectra enable fingerprinting, lineage attribution, and improved logit-based extraction, while suppressing them destroys inference utility.
Odin is the first open-source end-to-end GPU CKKS implementation of Llama-3, an FHE inference system that co-designs ciphertext packing and model execution for Llama and uses a feature-major cross-layer layout to unify residual connections and layer interfaces.
Yu-Hang Fan, Yu-Si Chen, Kan-Yu Ye et al.· arXiv.org· 1 citation
U NITED is presented, a shuffle-free MPSU framework based on a new data-structure abstraction, Oblivious Insert-Only In-vertible Sketches (OIIS), which ensures that each distinct union element is inserted exactly once while hiding the insertion condition and modified locations.
S. Trường, Nguyen, Lucas Piske et al.· 0 citations
This paper revisits the continuous noise sampling protocols and makes several improvements in both security and efficiency and turns to discrete sampling at the granularity of individual biased bits to address the security and efficiency issues together.
A setting in which the adversary uses only the public sampling-and-evaluation oracle available to honest participants, the induced distribution is arbitrary, and the goal is to attack future ciphertexts rather than recover the full algebraic span is studied, yielding a generic impossibility theorem.