# SILMARILS: Information-Theoretic and Quantum-Secure Designated-Verifier Signatures

Crypto.world

> Existing information-theoretic signature schemes offer unconditional security but generally lack efficient, reusable, simulation-secure constructions for multi-party, multi-use settings.

**arXiv**: 2605.03230

**Published**: 2026-05-04

**Authors**: Hassan Khodaiemehr, Khadijeh Bagheri, Chen Feng, Dariia Porechna

## AI summary

## Overview

**Research area:** Cryptography and security — specifically information-theoretic (IT) digital signatures, designated-verifier signatures, secure multiparty computation, and quantum-safe authentication for distributed systems such as blockchains.

**Technical level:** Advanced. The paper assumes familiarity with simulation-based security definitions, the real/ideal paradigm, secret sharing, and quantum-information concepts such as density operators, trace distance, and CPTP maps.

**Scope (one sentence):** The paper designs and proves the security of SILMARILS, a lightweight signature framework built from a minimal algebraic core over a finite field and perfect 2-out-of-2 Shamir secret sharing, offering a two-party transferable designated-verifier mode and a three-party information-theoretic mode, both analyzed in classical and quantum adversary models.

## What This Paper Is About

Existing information-theoretic signature schemes offer unconditional security but generally lack efficient, reusable, simulation-secure constructions for multi-party, multi-use settings. The authors build a signature framework, SILMARILS, that combines a very small algebraic core over the field of order p with perfect 2-out-of-2 Shamir secret sharing, and they prove its security in both a two-party transferable designated-verifier setting and a three-party broadcast setting. Their goal is not to beat standardized post-quantum cryptography on efficiency, but to show that IT techniques can deliver practical, compact authentication for lightweight designated-verifier use cases.

## Key Contributions

1. **A two-party transferable designated-verifier (TDV) signature scheme.** The designated verifier can simulate accepting transcripts that are indistinguishable from real ones (Jakobsson–Sako–Impagliazzo DV security), and may publish a receipt r that enables public verification without letting any external party distinguish signed from simulated transcripts. Since DV signatures permit simulation, standard EUF-CMA cannot hold for the designated verifier, so the authors instead prove EUF-CMA^¬DV security for all non-designated verifiers in both the random oracle model (ROM) and the quantum random oracle model (QROM).
2. **A three-party statistically secure signature protocol in the broadcast model of Fitzi et al.** The protocol achieves simulation-based security with error 1/p, analyzed in the Pure IT model, IT+ROM, and QROM.
3. **A quantum generalization of the Fitzi et al. characterization theorem.** The authors lift the classical result (Lemma 18.15 in Fitzi et al.) that reduces simulation-based security to four local properties — correctness, unforgeability, transferability, and secrecy — to the quantum setting with classical inputs and outputs, proving an O(ε) equivalence (Theorem 3.1). Crucially, the lifted proof avoids rewinding, forking, and oracle programming, which are the classical ROM techniques that fail in the QROM.
4. **An architecture proposal for post-quantum-safe blockchains.** SILMARILS is positioned to replace validator signatures during consensus and user transaction message authentication, using TLS-derived per-pair secrets to avoid large user-facing PQ signatures in high-frequency interactions, with commit–reveal mechanisms for randomness generation. The concrete ledger integration is deferred to a companion publication.

## Main Findings

- **Perfect secret sharing gives the scheme its quantum resilience.** Each share of the 2-out-of-2 Shamir sharing is uniformly distributed over F_p and independent of the secret, so the security is purely information-theoretic and does not rest on computational assumptions that quantum algorithms could break.
- **The designated verifier can produce indistinguishable forgeries by design.** This is the defining feature of DV signatures, and it is why the paper proves EUF-CMA^¬DV rather than full EUF-CMA: unforgeability is guaranteed only against parties other than the designated verifier.
- **Receipts preserve auditability without breaking deniability.** The verifier may publish a receipt r enabling public verification, yet even with r, no external party can tell whether a transcript was signed or simulated.
- **The three-party mode achieves error 1/p.** Under the Fitzi et al. broadcast model with three parties (signer P₁, holder P₂, verifier P₃) and a static, active adversary corrupting at most one party, the protocol is statistically secure with simulation error 1/p.
- **The 1/3 lower bound is circumvented by moving to a richer setup.** The paper cites the classical lower bound (Corollary 18.7 in Fitzi et al.) that any ε-secure three-party signature from scratch using only noiseless channels and no setup must satisfy ε ≥ 1/3; the construction avoids this by working in a richer setting with authenticated channels, pre-shared keys, or correlated randomness.
- **The quantum characterization theorem holds in both directions.** If a three-party protocol satisfies correctness, unforgeability, transferability, and secrecy with error at most ε, it is an O(ε)-secure implementation of the ideal functionality f_DS; conversely, any ε-secure implementation satisfies the four properties with error O(ε). The concrete proof bounding shows, for example, simulation error at most 2ε in the case where P₁ and P₂ are honest.
- **Keys and signatures are claimed to be very compact.** Table 1 classifies SILMARILS as having "very small" keys and "very small" signatures, based on signer–verifier shared randomness (TDV), with quantum security. The paper explicitly states that it does not aim to outperform post-quantum cryptography, and that a fair comparison against other designated-verifier schemes is omitted due to space and the complexity of aligning security models.
- **Quantitative efficiency benchmarks are not reported in the provided content.** The paper's Section 8 is described as an efficiency evaluation, but no concrete key sizes, signature sizes, timings, or parameter sets appear in the material available.

## Methodology in Plain English

The authors start from the observation that IT signatures are possible but awkward: existing constructions are either not simulation-secure, not reusable, or too inefficient for multi-party, multi-use settings. Rather than building a heavy new scheme, they take a deliberately minimal algebraic core over a finite field and add perfect 2-out-of-2 Shamir secret sharing as the structural backbone — a degree-1 polynomial f(x) = s + ax with two public interpolation points, reconstructed by Lagrange interpolation. Because every individual share is uniform and independent of the secret, the scheme inherits secrecy without computational assumptions.

They then split the design into two modes. In the two-party mode, the shared randomness is ephemeral and used only to enable controlled transferability, not to authenticate messages — this is what separates their construction from an ordinary MAC, which is symmetric and non-transferable. Security for everyone except the designated verifier is proved in the random oracle model and the quantum random oracle model. For the three-party mode, they plug the construction into the Fitzi et al. ideal functionality f_DS and its known reduction of simulation-based security to four checkable local properties, then prove each property holds with small error.

The methodological novelty is in the quantum step. Classical ROM proofs usually rely on rewinding, forking lemmas, or programming the oracle — none of which survive in the QROM. The authors note that the Fitzi et al. proof never interacts with the adversary algorithmically; it reasons only about joint distributions of inputs, outputs, and views. So they replace variational distance with trace distance, views with density operators, and classical coupling with the quantum coupling lemma, relying on the contractivity of CPTP maps and the triangle inequality. The result is a proof that lifts cleanly to quantum adversaries without any of the fragile techniques.

## Why This Matters

**Impact on research.** The paper connects two usually separate literatures: information-theoretic signature constructions and designated-verifier signatures. Its main theoretical contribution is showing that the Fitzi et al. simulation-based characterization of three-party signature security survives quantization, which gives future work a reusable, rewind-free template for proving QROM security of distributed signature protocols. It also sharpens the boundary between TDV signatures and MACs, arguing that ephemeral shared randomness for transferability is fundamentally different from a long-term symmetric key.

**Real-world applications:**

- **User–validator interactions in blockchains**, where a user and a validator can authenticate at high frequency using TLS-derived per-pair secrets, avoiding large post-quantum signatures on every message while receipts preserve public auditability.
- **Validator-to-validator consensus authentication**, which becomes cheaper and simpler, with randomness generated through lightweight commit–reveal.
- **Long-lived archival records**, where signatures must remain unforgeable decades later and information-theoretic (rather than assumption-based) security avoids reliance on hardness problems that quantum computers might break.
- **Constrained or high-throughput environments** needing constant-time verification and compact receipts with a small on-chain footprint.

**Industry relevance.** Standardized post-quantum schemes are, as the authors concede, already "more than sufficient" for most blockchain uses. The pitch here is narrower but real: where the on-chain footprint of PQ signatures is the bottleneck, or where end-to-end information-theoretic guarantees must not be broken by dropping in a computational signature inside an MPC protocol, SILMARILS offers a lightweight alternative with explicit, small error bounds.

## Future Directions

1. **A full ledger integration and performance study.** The authors explicitly defer concrete blockchain integration to a companion publication, so end-to-end measurements against standardized post-quantum schemes remain open.
2. **A fair comparison with other designated-verifier schemes.** This is stated as omitted due to space and the difficulty of aligning security models — a systematic comparison under a common model is a natural next step.
3. **Extending beyond one corruption in the three-party mode.** The current analysis is for at most one actively corrupted party; whether the framework extends to larger thresholds or richer adversarial structures is not addressed in the provided content.
4. **Practical instantiation guidance for the hash function across all three models.** The paper analyzes Pure IT, IT+ROM, and QROM, but the concrete choice of encoding or hash and the resulting parameter selection (values of p, weights, and security levels) are not specified in the available material.

## Target Audience

Cryptographers working on information-theoretic security, designated-verifier signatures, or quantum-safe protocol design will get the most from this paper, particularly those interested in QROM proof techniques that avoid rewinding. It also suits MPC and distributed-systems researchers who need signatures inside protocols where computational assumptions would break end-to-end IT guarantees. Blockchain protocol architects evaluating alternatives to standardized post-quantum signatures for high-frequency authentication are a secondary audience, though they should note that the ledger integration and efficiency benchmarks are not in this paper. Readers without a background in simulation-based security and quantum-information basics will find the security sections difficult.

## Authors’ abstract

SILMARILS is built from a minimal algebraic core over $\mathbb{F}_p$ using true randomness and perfect $2$-out-of-$2$ Shamir secret sharing. The framework supports both two-party and three-party modes. In the two-party setting, SILMARILS realizes a transferable designated-verifier (TDV) signature scheme. The designated verifier can simulate accepting transcripts indistinguishable from real ones, achieving Jakobsson-Sako-Impagliazzo DV security. The verifier may publish a…

[The full abstract on arXiv](https://arxiv.org/abs/2605.03230)

[Read the original paper](https://arxiv.org/abs/2605.03230)

---

Canonical: https://crypto.world/paper/silmarils-information-theoretic-and-quantum-secure-designated-verifier-2605.03230
