aiqbench analysis
The Safhire framework advances practical private inference by hybridizing Fully Homomorphic Encryption (FHE) with client-side plaintext processing to eliminate costly bootstrapping operations. This approach achieves significant latency reductions over existing FHE baselines while maintaining model confidentiality through randomized shuffling techniques. The work represents a critical step toward deploying secure, high-performance machine learning in cloud environments without compromising data privacy.
Practical and Private Hybrid ML Inference with Fully Homomorphic Encryption
arXiv:2509.01253v2 Announce Type: replace-cross Abstract: In contemporary cloud-based services, protecting users' sensitive data and ensuring the confidentiality of the server's model are critical. Fully homomorphic encryption (FHE) enables inference directly on encrypted inputs, but its practicality is hindered by expensive bootstrapping and inefficient approximations of non-linear activations. We introduce Safhire, a hybrid inference framework that executes linear layers under encryption on the server while offloading non- linearities to the client in plaintext. This design eliminates bootstrapping, supports exact activations, and significantly reduces computation cost. To safeguard model confidentiality despite client access to intermediate outputs, Safhire applies randomized shuffling, which obfuscates intermediate values and makes it practically impossible for a client to reconstruct the model. To further reduce inference latency, Safhire incorporates advanced optimizations such as fast ciphertext packing and partial extraction. Evaluations on multiple standard models and datasets show that Safhire achieves up to 7.6x and 7.4x lower inference latency than Orion and Concrete-ML, respectively, which are two state-of-the-art FHE baselines, with manageable communication overhead and comparable accuracy. These results underline the practicality of our hybrid FHE inference scheme.
Abstract:In contemporary cloud-based services, protecting users' sensitive data and ensuring the confidentiality of the server's model are critical. Fully homomorphic encryption (FHE) enables inference directly on encrypted inputs, but its practicality is hindered by expensive bootstrapping and inefficient approximations of non-linear activations. We introduce Safhire, a hybrid inference framework that executes linear layers under encryption on the server while offloading non- linearities to the client in plaintext. This design eliminates bootstrapping, supports exact activations, and significantly reduces computation cost. To safeguard model confidentiality despite client access to intermediate outputs, Safhire applies randomized shuffling, which obfuscates intermediate values and makes it practically impossible for a client to reconstruct the model. To further reduce inference latency, Safhire incorporates advanced optimizations such as fast ciphertext packing and partial extraction. Evaluations on multiple standard models and datasets show that Safhire achieves up to 7.6x and 7.4x lower inference latency than Orion and Concrete-ML, respectively, which are two state-of-the-art FHE baselines, with manageable communication overhead and comparable accuracy. These results underline the practicality of our hybrid FHE inference scheme.
Cryptography and Security (cs.CR); Machine Learning (cs.LG)
arXiv:2509.01253 [cs.CR]
(or arXiv:2509.01253v2 [cs.CR] for this version)
https://doi.org/10.48550/arXiv.2509.01253 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 270 words
Related coverage
A Solution to Cryptographic Boxes for Unfriendly AI
LessWrong · Jul 19, 2026
Lattice-Based Cryptography Explained
Quantum Zeitgeist · Aug 15, 2026
End-to-end encrypted ML inference with Amazon SageMaker AI and FHE
AWS · Jun 8, 2026
ATLAS: Automated Approximation of Transformers for Efficient Homomorphic Inference in One Hour
arXiv · Jul 28, 2026