# Quantum preprocessing for information-theoretic security in two-party computation

## TL;DR

This paper introduces quantum protocols for generating bipartite correlations that enable information-theoretic security in two-party computation, even with noise and aborts, advancing secure quantum and classical computing methods.

## Contribution

It proposes bipartite quantum protocols for approximately generating secure correlations without a third party, with methods to handle noise and achieve high success probability.

## Key findings

- Protocols can generate useful one-time tables with high probability in noiseless conditions.
- Security depends on noise levels but can be maintained with proposed noise mitigation methods.
- Generated correlations can implement no-signaling correlations like PR-boxes with communication.

## Abstract

In classical two-party computation, a trusted initializer who prepares certain initial correlations, known as one-time tables, can help make the inputs of both parties information-theoretically secure. We propose some bipartite quantum protocols with possible aborts for approximately generating such bipartite classical correlations with varying degrees of privacy, without introducing a third party. Under some weak requirements for the parties, the security level is nontrivial for use in bipartite computation. We show that the security is sometimes dependent on the noise level, but we propose a method for dealing with noise. The security is "forced security", which implies that the probability that some useful one-time tables are generated can approach $1$ in the noiseless case under quite weak assumptions about the parties, although the protocols allow aborts. We show how to use the generated one-time tables to achieve nontrivial information-theoretic security in generic two-party classical or quantum computation tasks, including (interactive) quantum homomorphic encryption. Our methods provide check-based implementations of some no-signaling correlations, including the PR-box type, with the help of communication which carry no information about the inputs in the generated correlations.