secure multi-party computation

为解决多方安全计算中跨领域分析问题，提出CryptDough引擎，通过层次化设计和虚拟向量技术，在保证隐私的同时提高性能。

Asymptotic security in secure multiparty computation (MPC) lacks quantifiable guarantees in practice. Method: We propose the first method for automatically computing concrete security bounds within the Universal Composability (UC) framework, inspired by IPDL-based modeling and implemented via the Maude logical rewriting system to formally capture the precise relationship between protocol execution time and adversary advantage. Contribution/Results: Our approach significantly simplifies simulation-based proofs and enables automated concrete security analysis. We conduct four case studies, including the first formal verification of concrete security bounds for the N-party GMW protocol—achieving a 72% reduction in proof size (567 lines of code) compared to prior work. This establishes the first concrete security analysis framework for MPC protocols that is simultaneously rigorous, computationally tractable, and scalable.

Prior work lacks a systematic analysis linking commitment scheme (CS) security properties to the robustness, privacy, and composable security of multi-party computation (MPC) protocols. Method: This paper establishes, for the first time, a formal mapping between CS types—including Pedersen, Fiat-Shamir, and universally composable (UC)-secure CS—and MPC construction paradigms (e.g., GMW, BGW, SPDZ), adversarial models, and application contexts. It proposes functional and security-driven CS selection criteria and conducts cryptographic analysis alongside cross-framework evaluation to quantify CS impact on MPC efficiency, resilience against adaptive attacks, and composability. Contribution/Results: The work bridges theoretical security guarantees with practical protocol design, yielding principled guidelines for CS instantiation in MPC. It provides both theoretical foundations and actionable insights for building robust, deployable privacy-preserving systems.

This work proposes a novel framework for cryptographic security by introducing Grothendieck topologies and sheaf theory into cryptography, thereby establishing a topos-theoretic foundation for security modeling. Departing from traditional game- or simulation-based definitions that lack a unified mathematical structure, the approach models an adversary’s observations as a Grothendieck site and protocol transcripts as sheaves. Crucially, it demonstrates that Σ-protocol transcripts form a torsor in the associated sheaf topos: local triviality corresponds to zero-knowledge, while the absence of global sections captures soundness. The efficacy of this framework is validated through the Schnorr protocol, illustrating how key security properties of cryptographic protocols can be uniformly characterized through categorical and geometric lenses.

This work addresses the challenge of simultaneously achieving weak-link tolerance and low computational overhead in high-throughput secure multi-party computation (MPC) over heterogeneous networks. We propose efficient ring-based three-party (semi-honest) and four-party (malicious, single-corruption) protocols. Our approach integrates ring-based secret sharing, lightweight arithmetic circuit optimization, and a high-throughput C++ implementation. While preserving optimal communication complexity—3 ring elements per multiplication for 3PC and 5 for 4PC—we achieve, for the first time, dual breakthroughs in weak-link robustness and computational efficiency: halving the number of gate instructions and improving throughput in heterogeneous networks by up to 8×. Experiments on a 25 Gbit/s LAN show our 5/6 protocol attains >10⁹ 32-bit multiplications/sec or 3.2×10¹⁰ AND gates/sec—outperforming mainstream frameworks such as MP-SPDZ by 2–3 orders of magnitude—and set new records for throughput in both 3PC and 4PC.

This work addresses the challenge of simultaneously ensuring parameter confidentiality and Byzantine robustness in large-scale decentralized learning. The authors propose Giskard, a novel protocol that, for the first time, achieves Byzantine-robust aggregation with sublinear communication complexity while preserving the privacy of model parameters. Giskard organizes participants into a logarithmic-depth committee tree and, within each committee, combines BGW-style secure multiparty computation with distributed binary search to perform coordinate-wise approximate median aggregation. Theoretical analysis establishes the protocol’s security and correctness, and empirical evaluation demonstrates that, at scales involving millions of participants, Giskard tolerates up to $n/4$ Byzantine nodes, substantially reduces communication overhead, and maintains high model utility.

Democracies are built upon secure and reliable voting systems. Electronic voting systems seek to replace ballot papers and boxes with computer hardware and software. Proposed electronic election schemes have been subjected to scrutiny, with researchers spotting inherent faults and weaknesses. Inspired by physical voting systems, we argue that any electronic voting system needs two essential properties: ballot secrecy and verifiability. These properties seemingly work against each other. An election scheme that is a complete black box offers ballot secrecy, but verification of the outcome is impossible. This challenge can be tackled using standard tools from modern cryptography, reaching a balance that delivers both properties. This tutorial makes these ideas accessible to readers outside electronic voting. We introduce fundamental concepts such as asymmetric and homomorphic encryption, which we use to describe a general electronic election scheme while keeping mathematical formalism minimal. We outline game-based cryptography, a standard approach in modern cryptography, and introduce notation for formulating elections as games. We then give precise definitions of ballot secrecy and verifiability in the framework of game-based cryptography. A principal aim is introducing modern research approaches to electronic voting.

This work addresses the lack of a composable, verifiable framework for mechanized cryptographic proofs that integrates with general-purpose mathematical libraries. It introduces HOPSCOTCH, the first game-hopping proof framework deeply integrated with Lean 4’s Mathlib, which models security definitions as indistinguishability between stateful probabilistic oracles via shallow embedding and employs state abstraction to enable flexible oracle transformations. The framework formalizes the structure of game-hopping reasoning, automates the derivation of computational soundness theorems, and successfully verifies the IND-CCA security of Encrypt-then-MAC, the security of ElGamal under the DDH assumption, the reduction from one-time secrecy to public-key IND-CPA, and the first mechanized proof of a non-constant-depth GGM pseudorandom function construction.

This work addresses the challenge of achieving privacy-preserving exact summation of Boolean bitstreams in multi-party settings while defending against inference attacks by both servers and aggregators. The authors propose PolyVeil, a secure aggregation protocol that encodes private bits as permutation matrices within the Birkhoff polytope, leveraging a two-tier architecture. Notably, PolyVeil operates without public-key infrastructure and uniquely integrates the geometric structure of the Birkhoff polytope with differential privacy. It outputs exact sums while guaranteeing perfect simulation-based security against the server and rendering likelihood inference for the aggregator #P-hard. The full variant achieves zero statistical distance security, whereas a compressed variant offers non-trivial ε-differential privacy under moderate signal-to-noise ratios, all with only O(k) communication complexity.

Decentralized grassroots platforms lack fault tolerance against faulty or malicious participants; existing systems cannot reliably achieve identity authentication, secure communication, and code integrity verification without secure programming support. Method: We propose a logic programming language tailored for grassroots platforms, integrating multi-agent modeling, concurrent semantics, and cryptographic primitives. It employs cryptographically bound logical variables to realize decentralized applications. Crucially, we design a novel pairwise single-read–single-write logical variable mechanism that simultaneously guarantees identity authentication and code integrity within a logic programming framework—first of its kind. Contribution/Results: We formally prove the mechanism’s computational security, acyclicity, and monotonicity. We implement a verifiable peer-to-peer protocol enabling smartphone-deployable grassroots social graphs and empirically validate end-to-end secure communication in real-world deployments.