Backend
Classical Verification of Quantum Computation Beyond the Polynomial Hierarchy
arXiv cs.CR0 views
arXiv:2610.37635v1 Announce Type: cross Abstract: It is not known whether a classical polynomial-time verifier can verify arbitrary polynomial-time quantum computations with an efficient quantum honest prover without computational assumptions. We show that, relative to a classical oracle, such verification is possible for a language beyond the polynomial hierarchy that is decided exactly in quantum polynomial time. The protocol is public-coin, with perfect completeness and soundness error at most 1/4. To our knowledge, no language with such a proof system was previously proved to lie outside AM relative to an oracle. For every unbounded polynomial-time computable function, the oracle can be chosen so that the message count and the oracle query counts of the quantum algorithm, verifier, and honest prover are bounded by this function of the input length. The message count is optimal up to a constant factor, since for a suitable oracle every interactive proof system for the language uses at least a constant fraction of the messages of our protocol on infinitely many lengths, with a constant depending on its verifier. The language is a variant of recursive Fourier sampling in which code tables encode secrets that address data tables. The tables are evaluated recursively, and the oracle stores only the leaves. The quantum algorithm recovers every secret exactly, and the verifier checks the prover's secrets recursively against random code entries. The recursion restricts the input in a correlated way, as the same secret fixes the code bits and chooses the free coordinate, so H{\aa}stad's switching lemma does not apply. We prove a switching theorem for these restrictions, which may be of independent interest. For instances of sufficient height, no constant-depth circuit of size $2^{\mathrm{poly}(n)}$ changes its output with probability above $2^{-\Omega(n)}$ when the answer of a random instance is flipped by changing a single leaf.
Source:https://arxiv.org/abs/2610.37635PreviousWhen Is a Relation Worth Trusting? Manipulation Costs, Exposure, and the Trust Trap in Multi-Relational Graph LearningNextHeads of Security: Learned Attention Gates Improve Prompt Injection Robustness
Related posts

Build headlines said 92x faster. I measured my own crate first.
0dev.to DevOps
PromQL Made Easier: Build, Understand, and Debug Prometheus Queries with AabiTech
0dev.to DevOps
Causely Feature Demo: Diagnosing Failures in an Agentic App
0dev.to DevOps

LLM Gateway & FinOps Architecture for B2B SaaS: Semantic Caching, Routing & Tenant Quotas
0dev.to DevOps

Leaked TLS Private Keys: Proving Compromise, Finding Every Affected Cert & Rotating Safely
0dev.to DevOps

Running Frontier AI Locally Still Doesn't Add Up
0dev.to DevOps