pith. sign in

arxiv: 2606.24379 · v1 · pith:7NRS6ZMBnew · submitted 2026-06-23 · 💻 cs.CR · cs.CL

ComputeFHE: A Privacy-Preserving General-Purpose Computation Library

Faris Serdar Tasel , Efe Ciftci This is my paper

Pith reviewed 2026-06-25 23:27 UTC · model grok-4.3

classification 💻 cs.CR cs.CL
keywords Fully Homomorphic EncryptionTFHEPrivacy-Preserving ComputationC++ LibraryBootstrapping OptimizationArithmetic Logic UnitEncrypted Data TypesOpenFHE
0
0 comments X

The pith

ComputeFHE supplies C++ encrypted data types and operations on TFHE that cut bootstrapping steps and reach up to 3.9x speedup on selected operations.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper presents ComputeFHE as an open-source C++ library built on OpenFHE that lets developers write privacy-preserving programs using familiar imperative code on encrypted integers and fixed-point values. It supplies arithmetic, logical, comparison, conditional, and oblivious array operations through both standard TFHE gate logic and an optimized ALU built from FHE-friendly primitives. Experiments reported in the paper show the optimized path reduces bootstrapping operations enough to deliver performance gains of up to 3.9 times on chosen workloads. A built-in simulation mode further lets users measure circuit cost and bootstrapping needs without running the actual cryptography. The library therefore aims to lower both the programming effort and the runtime overhead that have limited practical use of fully homomorphic encryption.

Core claim

ComputeFHE is a library that enables general-purpose computation on encrypted data by providing encrypted integer and fixed-point types together with a full set of arithmetic, logical, comparison, conditional, and oblivious array-access operations, implemented both through conventional two-input TFHE gates and through an optimized ALU architecture that uses FHE-friendly logic primitives to reduce the number of bootstrapping operations.

What carries the argument

The optimized Arithmetic Logic Unit (ALU) architecture utilizing FHE-friendly logic primitives, which replaces standard two-input gate circuits to lower the bootstrapping count for arithmetic and logical operations.

If this is right

  • Developers can express algorithms in ordinary imperative style while operating directly on encrypted data.
  • Selected arithmetic and logical operations require fewer bootstrapping steps than standard TFHE gate circuits.
  • A simulation mode supplies circuit-complexity and bootstrapping-cost estimates without executing cryptographic primitives.
  • Both the conventional gate-based path and the optimized ALU path can be used within the same library.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The simulation mode could support automated tools that rewrite circuits to further reduce bootstrapping before any encrypted execution occurs.
  • The library's C++ interface may allow direct reuse of existing non-encrypted codebases with minimal changes when privacy requirements arise.
  • Wider availability of such a library could shift experimental privacy-preserving work from specialized research groups toward ordinary software teams.

Load-bearing premise

The reported reductions in bootstrapping operations and the 3.9x performance gains depend on the particular workloads, hardware platform, and baseline implementations chosen for the experiments.

What would settle it

Public replication of the reported operations on documented workloads and hardware that measures either no reduction in bootstrapping count or a speedup materially below 3.9x.

Figures

Figures reproduced from arXiv: 2606.24379 by Efe Ciftci, Faris Serdar Tasel.

Figure 1
Figure 1. Figure 1: UML Diagram for ComputeFHE ComputeFHE utilizes the OpenFHE library to provide the fundamen￾tal TFHE functionalities, including key generation, encryption, decryp￾tion, and homomorphic computation. The OpenFHE library employs recommended cryptosystem parameters which corresponds to various security levels, such as 128-, 192-, and 256-bit security, supports multi￾ple bootstrapping techniques, and provides im… view at source ↗
Figure 2
Figure 2. Figure 2: Quickstart example. (a) Standard ALU implementation, (b) Optimized ALU [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Encrypted sort example. (a) Standard (b) Optimized [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
read the original abstract

Fully Homomorphic Encryption (FHE) enables computations to be performed directly on encrypted data while preserving data confidentiality. However, its practical applications remain limited by high computational costs and development complexity. This paper presents ComputeFHE, an open-source C++ library that facilitates the development of privacy-preserving applications based on the TFHE cryptosystem. The library provides encrypted integer and fixed-point data types together with arithmetic, logical, comparison, conditional, and oblivious array-access operations which allow developers to implement algorithms using a familiar imperative programming paradigm. ComputeFHE supports both conventional TFHE arithmetic based on standard two-input logic gates and an optimized Arithmetic Logic Unit (ALU) architecture utilizing FHE-friendly logic primitives. Experimental results demonstrate significant reductions in the number of required bootstrapping operations, achieving performance improvements of up to 3.9x for selected operations. In addition, the library includes a simulation mode that enables testing, debugging, and complexity analysis without performing actual cryptographic computations while providing circuit complexity and bootstrapping costs. Built on top of OpenFHE, ComputeFHE offers a practical and accessible framework for developing and evaluating privacy-preserving algorithms and applications.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 1 minor

Summary. The manuscript presents ComputeFHE, an open-source C++ library built on OpenFHE for TFHE-based fully homomorphic encryption. It offers encrypted integer and fixed-point types with arithmetic, logical, comparison, conditional, and oblivious array access operations. The library supports both conventional TFHE gate-based arithmetic and an optimized ALU architecture using FHE-friendly primitives, claiming reductions in bootstrapping operations and up to 3.9x performance improvements for selected operations. It also includes a simulation mode for testing and complexity analysis without actual computations.

Significance. If the performance claims are substantiated with reproducible experiments, the library could provide a practical tool for developers to implement privacy-preserving algorithms using an imperative paradigm, potentially reducing development complexity in FHE applications. The dual architecture and simulation mode are notable features. However, the current presentation of results limits the ability to assess the significance of the optimizations.

major comments (1)
  1. [Abstract] Abstract: The claim of 'significant reductions in the number of required bootstrapping operations, achieving performance improvements of up to 3.9x for selected operations' supplies no workload definitions, hardware platform, baseline (library's own conventional path or external), input sizes, OpenFHE version, or measurement protocol. Without these, the quantitative result cannot be verified and is load-bearing for the central claim that the optimized ALU yields the reported gains.
minor comments (1)
  1. [Abstract] Abstract: The qualifier 'selected operations' is imprecise; the manuscript should enumerate the tested operations and their input sizes to permit evaluation of whether they are representative.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive feedback. We address the single major comment below and agree that revisions are needed to strengthen verifiability of the performance claims.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The claim of 'significant reductions in the number of required bootstrapping operations, achieving performance improvements of up to 3.9x for selected operations' supplies no workload definitions, hardware platform, baseline (library's own conventional path or external), input sizes, OpenFHE version, or measurement protocol. Without these, the quantitative result cannot be verified and is load-bearing for the central claim that the optimized ALU yields the reported gains.

    Authors: We agree that the abstract, standing alone, omits the experimental parameters needed to assess the 3.9x claim. The full manuscript provides these details in the Experimental Evaluation section (workloads consisting of selected arithmetic, logical, and comparison operations on encrypted integers and fixed-point values; hardware platform; baseline as the library's own conventional TFHE gate-based path; input sizes; OpenFHE version; and timing protocol). To make the central claim verifiable without requiring the reader to consult the body, we will revise the abstract to concisely state the key experimental conditions and workloads used for the reported speedup. revision: yes

Circularity Check

0 steps flagged

No circularity; library implementation with empirical measurements only

full rationale

The paper describes an open-source C++ library built on OpenFHE that supplies encrypted data types and operations, plus a simulation mode. The sole quantitative claim is an experimental observation of up to 3.9x speedup and reduced bootstrapping counts for selected operations. No equations, fitted parameters, predictions, or self-citations appear in the abstract or described content. The reported speedups are direct measurements of the implemented artifact rather than quantities derived from any internal definition or prior self-citation, so the contribution remains self-contained and non-circular.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The library rests on the standard security assumptions of the TFHE cryptosystem (LWE hardness) and the correctness of the underlying OpenFHE implementation; no new free parameters, ad-hoc axioms, or invented entities are introduced in the abstract.

axioms (2)
  • standard math TFHE cryptosystem security reduces to the hardness of the Learning With Errors problem
    Invoked implicitly by any TFHE-based construction; stated as background in the abstract's opening sentence.
  • domain assumption OpenFHE provides a correct and efficient implementation of TFHE primitives
    The library is built directly on OpenFHE; correctness of the new layer inherits from this dependency.

pith-pipeline@v0.9.1-grok · 5732 in / 1371 out tokens · 17188 ms · 2026-06-25T23:27:37.548645+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

17 extracted references · 10 canonical work pages

  1. [1]

    T. V. T. Doan, M.-L. Messai, G. Gavin, J. Darmont, A survey on implementations of homomorphic encryption schemes: Dt van thao et al., The Journal of Supercomputing 79 (13) (2023) 15098–15139. doi:10.1007/s11227-023-05233-z

    work page doi:10.1007/s11227-023-05233-z 2023

  2. [2]

    Fully homomorphic encryption using ideal lattices , booktitle =

    C. Gentry, Fully homomorphic encryption using ideal lattices, in: Pro- ceedings of the forty-first annual ACM symposium on Theory of com- puting, 2009, pp. 169–178.doi:10.1145/1536414.1536440

    work page doi:10.1145/1536414.1536440 2009

  3. [3]

    Chillotti, N

    I. Chillotti, N. Gama, M. Georgieva, M. Izabachene, Faster fully ho- momorphic encryption: Bootstrapping in less than 0.1 seconds, in: international conference on the theory and application of cryptology and information security, Springer, 2016, pp. 3–33.doi:10.1007/ 978-3-662-53887-6_1

    2016

  4. [4]

    Chillotti, N

    I. Chillotti, N. Gama, M. Georgieva, M. Izabachène, Faster packed homomorphic operations and efficient circuit bootstrapping for tfhe, in: International Conference on the Theory and Application of Cryp- tology and Information Security, Springer, 2017, pp. 377–408.doi: 10.1007/978-3-319-70694-8_14

    work page doi:10.1007/978-3-319-70694-8_14 2017

  5. [5]

    J. H. Cheon, K. Han, A. Kim, M. Kim, Y. Song, A full rns variant of approximate homomorphic encryption, in: International Conference on Selected Areas in Cryptography, Springer, 2018, pp. 347–368.doi: 10.1007/978-3-030-10970-7_16

    work page doi:10.1007/978-3-030-10970-7_16 2018

  6. [6]

    doi:10.1109/ACCESS.2024.3461729

    L.Bai, L.Bai, Y.Li, Z.Li, Researchonnoisemanagementtechnologyfor fully homomorphic encryption, IEEE Access 12 (2024) 135564–135576. doi:10.1109/ACCESS.2024.3461729

    work page doi:10.1109/access.2024.3461729 2024

  7. [7]

    Bergerat, J.-B

    L. Bergerat, J.-B. Orfila, A. Roux-Langlois, S. Tap, Accelerating tfhe with sorted bootstrapping techniques, in: International Conference on the Theory and Application of Cryptology and Information Security, Springer, 2025, pp. 68–100.doi:10.1007/978-981-95-5122-4_3

    work page doi:10.1007/978-981-95-5122-4_3 2025

  8. [8]

    Xiang, J

    B. Xiang, J. Zhang, Y. Deng, Y. Dai, D. Feng, Fast blind rotation for bootstrapping fhes, in: Annual International Cryptology Conference, Springer, 2023, pp. 3–36.doi:10.1007/978-3-031-38551-3_1

    work page doi:10.1007/978-3-031-38551-3_1 2023

  9. [9]

    Chillotti, D

    I. Chillotti, D. Ligier, J.-B. Orfila, S. Tap, Improved programmable bootstrapping with larger precision and efficient arithmetic circuits for tfhe, in: International Conference on the Theory and Application of 15 Cryptology and Information Security, Springer, 2021, pp. 670–699.doi: 10.1007/978-3-030-92078-4_23

    work page doi:10.1007/978-3-030-92078-4_23 2021

  10. [10]

    Zama, Concrete: TFHE Compiler that converts python programs into FHE equivalent,https://github.com/zama-ai/concrete(2022)

    2022

  11. [11]

    com/zama-ai/concrete-ml(2022)

    Zama, Concrete ML: a privacy-preserving machine learning library us- ing fully homomorphic encryption for data scientists,https://github. com/zama-ai/concrete-ml(2022)

    2022

  12. [12]

    Zama, TFHE-rs: A Pure Rust Implementation of the TFHE Scheme for Boolean and Integer Arithmetics Over Encrypted Data,https:// github.com/zama-ai/tfhe-rs(2022)

    2022

  13. [13]

    Gouert, D

    C. Gouert, D. Mouris, N. G. Tsoutsos, Helm: Navigating homomor- phic encryption through gates and lookup tables, IEEE Transactions on Information Forensics and Security (2025).doi:10.1109/TIFS.2025. 3544066

    work page doi:10.1109/tifs.2025 2025

  14. [14]

    Chaturvedi, A

    B. Chaturvedi, A. Chatterjee, A. Chattopadhyay, D. Mukhopadhyay, Bolt: Bootstrapping-aware logic resynthesis and technology mapping for efficient tfhe circuits, Cryptology ePrint Archive (2026)

    2026

  15. [15]

    F. S. Taşel, A. N. Saran, Improved arithmetic efficiency in tfhe through gate-level optimizations: Fs taşel, an saran, The Journal of Supercom- puting 81 (18) (2025) 1633.doi:10.1007/s11227-025-08107-8

    work page doi:10.1007/s11227-025-08107-8 2025

  16. [16]

    Boemer, S

    F. Boemer, S. Kim, G. Seifu, F. D. de Souza, V. Gopal, et al., In- tel HEXL (release 1.2),https://github.com/intel/hexl(september 2021)

    2021

  17. [17]

    Micciancio, Y

    D. Micciancio, Y. Polyakov, Bootstrapping in fhew-like cryptosystems, in: Proceedings of the 9th on Workshop on Encrypted Computing & Applied Homomorphic Cryptography, 2021, pp. 17–28.doi:10.1145/ 3474366.3486924. 16

    arXiv 2021