Encrypt What Matters: When Selective Homomorphic Inference Is Efficient

• URL: http://arxiv.org/abs/2609.09357v1
• Date: Tue, 08 Sep 2026 18:46:05 GMT
• Title: Encrypt What Matters: When Selective Homomorphic Inference Is Efficient
• Abstract summary: homomorphic encryption (FHE) enables inference on private data without revealing it to the server, but evaluating an entire input under FHE is expensive.<n>We study emphselective homomorphic inference, where only a sensitive region of interest is encrypted.<n>We find that locality is the key architectural property governing the benefit of selective homomorphic inference.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Fully homomorphic encryption (FHE) enables inference on private data without revealing it to the server, but evaluating an entire input under FHE is expensive. We study \emph{selective homomorphic inference}, where only a sensitive region of interest (ROI) is encrypted, and computations independent of that region are performed in plaintext. Selective evaluation produces the same output as full FHE on the same model, without retraining. Its efficiency depends on how quickly encrypted dependencies spread through the network. For small encrypted ROIs, locality-preserving architectures can achieve order-of-magnitude homomorphic-evaluation speedups, whereas architectures with early global mixing provide essentially no speedup. These results identify locality as the key architectural property governing the benefit of selective homomorphic inference.

Related papers

• GoldenRetriever: Non-Interactive Homomorphic Encrypted Retrieval for Privacy-Preserving RAG [20.647970767787225]
  We propose a non-interactive encrypted retrieval framework for RAG based on threshold selection.<n>Instead of performing expensive top-$k$ ranking under encryption, our approach selects documents whose similarity scores exceed a predefined threshold.<n>Results highlight threshold-based selection as a practical foundation for scalable and secure RAG systems.
  arXiv  Detail & Related papers  (2026-07-31T04:44:24Z)
• Tilewise Domain-Separated Selective Encryption for Remote Sensing Imagery under Chosen-Plaintext Attacks [3.4267780523755995]
  This paper proposes Tilewise Domain-Separated Selective Encryption ( TDS-SE), where per-tile keys are derived from a master secret via HKDF with explicit domain separation.<n> Structural leakage is evaluated using two reconstruction-based distinguishers -- a linear model and a lightweight convolutional neural network.<n>Cross-sample generalization exhibits mixed behavior across settings, consistent with an empirical evaluation perspective.
  arXiv  Detail & Related papers  (2026-02-27T07:57:31Z)
• Improving LLM Reasoning with Homophily-aware Structural and Semantic Text-Attributed Graph Compression [55.51959317490934]
  Large language models (LLMs) have demonstrated promising capabilities in Text-Attributed Graph (TAG) understanding.<n>We argue that graphs inherently contain rich structural and semantic information, and that their effective exploitation can unlock potential gains in LLMs reasoning performance.<n>We propose Homophily-aware Structural and Semantic Compression for LLMs (HS2C), a framework centered on exploiting graph homophily.
  arXiv  Detail & Related papers  (2026-01-13T03:35:18Z)
• UAGLNet: Uncertainty-Aggregated Global-Local Fusion Network with Cooperative CNN-Transformer for Building Extraction [83.48950950780554]
  Building extraction from remote sensing images is a challenging task due to the complex structure variations of buildings.<n>Existing methods employ convolutional or self-attention blocks to capture the multi-scale features in the segmentation models.<n>We present an Uncertainty-Aggregated Global-Local Fusion Network (UAGLNet) to exploit high-quality global-local visual semantics.
  arXiv  Detail & Related papers  (2025-12-15T02:59:16Z)
• What matters for Representation Alignment: Global Information or Spatial Structure? [64.67092609921816]
  Representation alignment (REPA) guides generative training by distilling representations from a strong, pretrained vision encoder to intermediate diffusion features.<n>We investigate a fundamental question: what aspect of the target representation matters for generation, its textitglobal revisionsemantic information.<n>We replace the standard projection layer in REPA with a simple convolution layer and introduce a spatial normalization layer for the external representation.
  arXiv  Detail & Related papers  (2025-12-11T16:39:53Z)
• Plaintext Structure Vulnerability: Robust Cipher Identification via a Distributional Randomness Fingerprint Feature Extractor [23.713094083283334]
  We present a method that does not learn end-to-end from ciphertext bytes.<n>Specifically, this method is based on a set of statistical tests to compute the randomness feature of the ciphertext.<n>The experimental results demonstrate that our method achieves high discriminative performance.
  arXiv  Detail & Related papers  (2025-11-11T14:29:42Z)
• Cryptanalysis on Lightweight Verifiable Homomorphic Encryption [8.478323897416617]
  Verifiable Homomorphic Encryption (VHE) is a cryptographic technique that integrates Homomorphic Encryption (HE) with Verifiable Computation (VC)<n>It serves as a crucial technology for ensuring both privacy and integrity in outsourced computation.<n>This paper presents efficient attacks that exploit the homomorphic properties of encryption schemes.
  arXiv  Detail & Related papers  (2025-02-18T08:13:10Z)
• CPR++: Object Localization via Single Coarse Point Supervision [55.8671776333499]
  coarse point refinement (CPR) is first attempt to alleviate semantic variance from an algorithmic perspective. CPR reduces semantic variance by selecting a semantic centre point in a neighbourhood region to replace the initial annotated point. CPR++ can obtain scale information and further reduce the semantic variance in a global region.
  arXiv  Detail & Related papers  (2024-01-30T17:38:48Z)
• When approximate design for fast homomorphic computation provides differential privacy guarantees [0.08399688944263842]
  Differential privacy (DP) and cryptographic primitives are popular countermeasures against privacy attacks. In this paper, we design SHIELD, a probabilistic approximation algorithm for the argmax operator. Even if SHIELD could have other applications, we here focus on one setting and seamlessly integrate it in the SPEED collaborative training framework.
  arXiv  Detail & Related papers  (2023-04-06T09:38:01Z)
• Confident Sinkhorn Allocation for Pseudo-Labeling [40.883130133661304]
  Semi-supervised learning is a critical tool in reducing machine learning's dependence on labeled data. This paper studies theoretically the role of uncertainty to pseudo-labeling and proposes Confident Sinkhorn Allocation (CSA) CSA identifies the best pseudo-label allocation via optimal transport to only samples with high confidence scores.
  arXiv  Detail & Related papers  (2022-06-13T02:16:26Z)
• Feature Completion for Occluded Person Re-Identification [138.5671859358049]
  RFC block can recover semantics of occluded regions in feature space. SRFC exploits the long-range spatial contexts from non-occluded regions to predict the features of occluded regions. TRFC module captures the long-term temporal contexts to refine the prediction of SRFC.
  arXiv  Detail & Related papers  (2021-06-24T02:40:40Z)
• Self-Challenging Improves Cross-Domain Generalization [81.99554996975372]
  Convolutional Neural Networks (CNN) conduct image classification by activating dominant features that correlated with labels. We introduce a simple training, Self-Challenging Representation (RSC), that significantly improves the generalization of CNN to the out-of-domain data. RSC iteratively challenges the dominant features activated on the training data, and forces the network to activate remaining features that correlates with labels.
  arXiv  Detail & Related papers  (2020-07-05T21:42:26Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.