International Association
for Cryptologic Research

Event date: 30 January to 3 February 2023

In this paper, we introduce a second-order masking of the AES using the minimal number of shares and a total of 1268 bits of randomness including the sharing of the plaintext and key. The masking of the S-box is based on the tower field decomposition of the inversion over bytes where the changing of the guards technique is used in order to re-mask the middle branch of the decomposition. The sharing of the S-box is carefully crafted such that it achieves first-order probing security without the use of randomness and such that the sharing of its output is uniform. Multi-round security is achieved by re-masking the state where we use a theoretical analysis based on the propagation of probed information to reduce the demand for fresh randomness per round. The result is a second-order masked AES which competes with the state-of-the-art in terms of latency and area, but reduces the randomness complexity over eight times over the previous known works. In addition to the corresponding theoretical analysis and proofs for the security of our masked design, it has been implemented on FPGA and evaluated via lab analysis.

The notion of distributed authenticated encryption was formally introduced by Agrawal et al. in ACM CCS 2018. In their work, they propose the DiSE construction building upon a distributed PRF (DPRF), a commitment scheme and a PRG. We show that most of their constructions do not meet some of the claimed security guarantees. In fact, all the concrete instantiations of DiSE, as well as multiple follow-up papers (one accepted at ACM CCS 2021), fail to satisfy their strongly-secure definitions. We give simple fixes for these constructions and prove their security. We also propose a new construction DiAE using an encryptment instead of a commitment. This modification dispenses with the need to buffer the entire message throughout the encryption protocol, which in turn enables implementations with constant RAM footprint and online message encryption. This is particularly interesting for constrained IoT devices. Finally, we implement and benchmark DiAE and show that it performs similarly to the original DiSE construction.

The question whether one way functions (i.e., functions that are easy to compute but hard to invert) exist is arguably one of the central problems in complexity theory, both from theoretical and practical aspects. While proving that such functions exist could be hard, there were quite a few attempts to provide functions which are one way "in practice", namely, they are easy to compute, but there are no known polynomial time algorithms {that} compute their (generalized) inverse (or that computing their inverse is as hard as notoriously difficult tasks, like factoring very large integers).

In this paper we study a different approach. We provide a simple heuristic, called self masking, which converts a given polynomial time computable function $f$ into a self masked version ${f}$, which satisfies the following: for a random input $x$, ${f}^{-1}({f}(x))=f^{-1}(f(x))$ w.h.p., but a part of $f(x)$, which is essential for computing $f^{-1}(f(x))$ is {\it masked} in ${f}(x)$. Intuitively, this masking makes it hard to convert an efficient algorithm which computes $f^{-1}$ to an efficient algorithm which computes ${f}^{-1}$, since the masked parts are available to $f$ but not to ${f}$. We apply this technique on variants of the subset sum problem which were studied in the context of one way functions, and obtain functions which, to the best of our knowledge, cannot be inverted in polynomial time by published techniques.

This note describes an observation discovered during a failed cryptanalysis attempt.

Let $P(x,y)$ be a bivariate polynomial with coefficients in $\mathbb{C}$. Form the $n\times n$ matrices $L(n)$ whose elements are defined by $P(i,j)$. Define the matrices $M(n)=L(n)-\mbox{ID}_n$.

It appears that $\mu(n)=(-1)^n\det(M_n)$ is a polynomial in $n$ that we did not characterize.

We provide a numerical example.

One of the most fundamental results in game theory is that every finite strategic game has a Nash equilibrium, an assignment of (randomized) strategies to players with the stability property that no individual player can benefit from deviating from the assigned strategy. It is not known how to efficiently compute such a Nash equilibrium --- the computational complexity of this task is characterized by the class PPAD, but the relation of PPAD to other problems and well-known complexity classes is not precisely understood. In recent years there has been mounting evidence, based on cryptographic tools and techniques, showing the hardness of PPAD. We continue this line of research by showing that PPAD is as hard as learning with errors (LWE) and the iterated squaring (IS) problem, two standard problems in cryptography. Our work improves over prior hardness results that relied either on (1) sub-exponential assumptions, or (2) relied on ``obfustopia,'' which can currently be based on a particular combination of three assumptions. Our work additionally establishes public-coin hardness for PPAD (computational hardness for a publicly sampleable distribution of instances) that seems out of reach of the obfustopia approach. Following the work of Choudhuri et al. (STOC 2019) and subsequent works, our hardness result is obtained by constructing an unambiguous and incrementally-updatable succinct non-interactive argument for IS, whose soundness relies on polynomial hardness of LWE. The result also implies a verifiable delay function with unique proofs, which may be of independent interest.

Modern SVD computation dates back to work in the 1960s that proposed the basis for the eigensystem package and linear algebra package routines. As a result of a long history of research, SVD is now widely applied in various scenarios, such as recommendation system and principal component analysis. Furthermore, federated SVD has emerged as a prevalent privacy-preserving technique. For example, the raw data are not required to be exchanged among different parties; instead, each party trains and processes locally and shares intermediate result. In general, there are two main categories: SVD over horizontally and vertically partitioned data. Imagine a dataset matrix M, where each row stands for a record from a data subject, and the columns stand for the attributes/features of the records. In the horizontally partitioned setting, each party holds a disjoint subset of the rows of M. While in the vertically partitioned setting, each party has a disjoint subset of the columns of M for all the rows. In real-world applications, the horizontally partitioned setting is much more common than the vertically partitioned setting. In this paper, we have proposed a privacy-preserving federated SVD scheme with secure aggregation. The proposed scheme can aggregate SVD results (eigenspace) from different devices and synchronise the aggregation result with all devices while maintaining privacy protection.

The current gold standard of cryptographic software is to write efficient libraries with systematic protections against timing attacks. In order to meet this goal, cryptographic engineers increasingly use high-assurance cryptography tools. These tools guide programmers and provide rigorous guarantees that can be verified independently by library users. However, high-assurance tools reason about overly simple execution models that elide micro-architectural leakage. Thus, implementations validated by high-assurance cryptography tools remain potentially vulnerable to micro-architectural attacks such as Spectre or Meltdown. Moreover, proposed countermeasures are not used in practice due to performance overhead.

We propose, analyze, implement and evaluate an approach for writing efficient cryptographic implementations that are protected against Spectre v1 attacks. Our approach ensures speculative constant-time, an information flow property which guarantees that programs are protected against Spectre v1. Speculative constant-time is enforced by means of a (value-dependent) information flow type system. The type system tracks security levels depending on whether execution is misspeculating. We implement our approach in the Jasmin framework for high assurance cryptography, and use it for protecting all implementations of an experimental cryptographic library that includes highly optimized implementations of symmetric primitives, of elliptic-curve cryptography, and of Kyber, a lattice-based KEM recently selected by NIST for standardization. The performance impact of our protections is very low; for example, less than 1% for Kyber and essentially zero for X25519.

In recent years, functional encryption (FE) has established itself as one of the fundamental primitives in cryptography. The choice of model of computation to represent the functions associated with the functional keys plays a critical role in the complexity of the algorithms of an FE scheme. Historically, the functions are represented as circuits. However, this results in the decryption time of the FE scheme growing proportional to not only the worst case running time of the function but also the size of the input, which in many applications can be quite large.

In this work, we present the first construction of a public-key collusion-resistant FE scheme, where the functions, associated with the keys, are represented as random access machines (RAMs). We base the security of our construction on the existence of: (i) public-key collusion- resistant FE for circuits and, (ii) public-key doubly-efficient private-information retrieval [Boyle et al., Canetti et al., TCC 2017]. Our scheme enjoys many nice efficiency properties, including input-specific decryption time.

We also show how to achieve FE for RAMs in the bounded-key setting with weaker efficiency guarantees from laconic oblivious transfer, which can be based on standard cryptographic assumptions. En route to achieving our result, we present conceptually simpler constructions of succinct garbling for RAMs [Canetti et al., Chen et al., ITCS 2016] from weaker assumptions.

In this paper, we follow the line of existing study on cryptographic enforcement of Role-Based Access Control (RBAC). Inspired by the study of the relation between the existing security definitions for such system, we identify two different types of attacks which cannot be captured by the existing ones. Therefore, we propose two new security definitions towards the goal of appropriately modelling cryptographic enforcement of Role-Based Access Control policies and study the relation between our new definitions and the existing ones. In addition, we show that the cost of supporting dynamic policy update is inherently expensive by presenting two lower bounds for such systems which guarantee correctness and secure access.

In most homomorphic encryption schemes based on the RLWE, the native plaintexts are represented as polynomials in a ring $Z_t[x]/x^N+1$ where $t$ is a plaintext modulus and $x^N+1$ is a cyclotomic polynomial with degree power of two. An encoding scheme should be used to transform some natural data types(such as integers and rational numbers) into polynomials in the ring. After a homomorphic computation on the polynomial is finished, the decoding procedure is invoked to obtain the result. However, conditions for decoding correctly are strict in a way. For example, the overflows of computation modulo both the plaintext modulus $t$ and the cyclotomic polynomial $x^N+1$ will result in a unexpected result for decoding. The reason is that decoding the part which is discarded by modular reduction is not 0. We combine number theory transformation with Hensel Codes to construct a scheme. Intuitively, decoding the discarded part will yield 0 so the limitations are overcome naturally in our scheme. On the other hand, rational numbers can be handled with high precision in parallel.

Broadcast is an essential primitive for secure computation. We focus in this paper on optimal resilience (i.e., when the number of corrupted parties $t$ is less than a third of the computing parties $n$), and with no setup or cryptographic assumptions.

While broadcast with worst case $t$ rounds is impossible, it has been shown [Feldman and Micali STOC'88, Katz and Koo CRYPTO'06] how to construct protocols with expected constant number of rounds in the private channel model. However, those constructions have large communication complexity, specifically $\mathcal{O}(n^2L+n^6\log n)$ expected number of bits transmitted for broadcasting a message of length $L$. This leads to a significant communication blowup in secure computation protocols in this setting.

In this paper, we substantially improve the communication complexity of broadcast in constant expected time. Specifically, the expected communication complexity of our protocol is $\mathcal{O}(nL+n^4\log n)$. For messages of length $L=\Omega(n^3 \log n)$, our broadcast has no asymptotic overhead (up to expectation), as each party has to send or receive $\mathcal{O}(n^3 \log n)$ bits. We also consider parallel broadcast, where $n$ parties wish to broadcast $L$ bit messages in parallel. Our protocol has no asymptotic overhead for $L=\Omega(n^2\log n)$, which is a common communication pattern in perfectly secure MPC protocols. For instance, it is common that all parties share their inputs simultaneously at the same round, and verifiable secret sharing protocols require the dealer to broadcast a total of $\mathcal{O}(n^2\log n)$ bits.

As an independent interest, our broadcast is achieved by a packed verifiable secret sharing, a new notion that we introduce. We show a protocol that verifies $\mathcal{O}(n)$ secrets simultaneously with the same cost of verifying just a single secret. This improves by a factor of $n$ the state-of-the-art.

Ring signatures allow a user to sign messages on behalf of an ad hoc set of users - a ring - while hiding her identity. The original motivation for ring signatures was whistleblowing [Rivest et al. ASIACRYPT'01]: a high government employee can anonymously leak sensitive information while certifying that it comes from a reliable source, namely by signing the leak. However, essentially all known ring signature schemes require the members of the ring to publish a structured verification key that is compatible with the scheme. This creates somewhat of a paradox since, if a user does not want to be framed for whistleblowing, they will stay clear of signature schemes that support ring signatures. In this work, we formalize the concept of universal ring signatures (URS). A URS enables a user to issue a ring signature with respect to a ring of users, independently of the signature schemes they are using. In particular, none of the verification keys in the ring need to come from the same scheme. Thus, in principle, URS presents an effective solution for whistleblowing.

The main goal of this work is to study the feasibility of URS, especially in the standard model (i.e. no random oracles or common reference strings). We present several constructions of URS, offering different trade-offs between assumptions required, the level of security achieved, and the size of signatures: * Our first construction is based on superpolynomial hardness assumptions of standard primitives. It achieves compact signatures. That means the size of a signature depends only logarithmically on the size of the ring and on the number of signature schemes involved. * We then proceed to study the feasibility of constructing URS from standard polynomially-hard assumptions only. We construct a non-compact URS from witness encryption and additional standard assumptions. * Finally, we show how to modify the non-compact construction into a compact one by relying on indistinguishability obfuscation.

Key Transparency (KT) systems allow end-to-end encrypted service providers (messaging, calls, etc.) to maintain an auditable directory of their users’ public keys, producing proofs that all participants have a consistent view of those keys, and allowing each user to check updates to their own keys. KT has lately received a lot of attention, in particular its privacy preserving variants, which also ensure that users and auditors do not learn anything beyond what is necessary to use the service and keep the service provider accountable.

Abstractly, the problem of building such systems reduces to constructing so-called append-only Zero-Knowledge Sets (aZKS). Unfortunately, existing aZKS (and KT) solutions do not allow to adequately restore the privacy guarantees after a server compromise, a form of Post-Compromise Security (PCS), while maintaining the auditability properties. In this work we address this concern through the formalization of an extension of aZKS called Rotatable ZKS (RZKS). In addition to providing PCS, our notion of RZKS has several other attractive features, such as a stronger (extractable) soundness notion, and the ability for a communication party with out-of-date data to efficiently “catch up” to the current epoch while ensuring that the server did not erase any of the past data.

Of independent interest, we also introduce a new primitive called a Rotatable Verifiable Random Function (VRF), and show how to build RZKS in a modular fashion from a rotatable VRF, ordered accumulator, and append-only vector commitment schemes.

We revisit the well-studied problem of preventing steganographic communication in multi-party communications. While this is known to be a provably impossible task, we propose a new model that allows circumventing this impossibility. In our model, the parties first publish a single message during an honest non-interactive pre-processing phase and then later interact in an execution phase. We show that in this model, it is indeed possible to prevent any steganographic communication in zero-knowledge protocols. Our solutions rely on standard cryptographic assumptions.

This paper studies Batch Private Information Retrieval (BatchPIR), a variant of private information retrieval (PIR) where the client wants to retrieve multiple entries from the server in one batch. BatchPIR matches the use case of many practical applications and holds the potential for substantial efficiency improvements over PIR in terms of amortized cost per query. Existing BatchPIR schemes have achieved decent computation efficiency but have not been able to improve communication efficiency at all. In this paper, we present the first BatchPIR protocol that is efficient in both computation and communication for a variety of database configurations. Specifically, to retrieve a batch of 256 entries from a database with one million entries of 256 bytes each, the communication cost of our scheme is 7.2$\sim$75x better than state-of-the-art solutions.

We investigate the relationship between the classical RSA and factoring problems when preprocessing is considered. In such a model, adversaries can use an unbounded amount of precomputation to produce an "advice" string to then use during the online phase, when a problem instance becomes known. Previous work (e.g., [Bernstein, Lange ASIACRYPT '13]) has shown that preprocessing attacks significantly improve the runtime of the best-known factoring algorithms. Due to these improvements, we ask whether the relationship between factoring and RSA fundamentally changes when preprocessing is allowed. Specifically, we investigate whether there is a superpolynomial gap between the runtime of the best attack on RSA with preprocessing and on factoring with preprocessing.

Our main result rules this out with respect to algorithms in a natural adaptation of the generic ring model to the preprocessing setting. In particular, in this setting we show the existence of a factoring algorithm (albeit in the random oracle model) with polynomially related parameters, for any setting of RSA parameters.

We provide a strong definition for committing authenticated-encryption (cAE), as well as a framework that encompasses earlier and weaker definitions. The framework attends not only to what is committed but also the extent to which the adversary knows or controls keys. We slot into our framework strengthened cAE-attacks on GCM and OCB. Our main result is a simple and efficient construction, CTX, that makes a nonce-based AE (nAE) scheme committing. The transformed scheme achieves the strongest security notion in our framework. Just the same, the added computational cost (on top of the nAE scheme's cost) is a single hash over a short string, a cost independent of the plaintext's length. And there is no increase in ciphertext length compared to the base nAE scheme. That such a thing is possible, let alone easy, upends the (incorrect) intuition that you can't commit to a plaintext or ciphertext without hashing one or the other. And it motivates a simple and practical tweak to AE-schemes to make them committing.

We address three main open problems concerning the use of radical isogenies, as presented by Castryck, Decru and Vercauteren at Asiacrypt 2020, in the computation of long chains of isogenies of fixed, small degree between elliptic curves over finite fields. Firstly, we present an interpolation method for finding radical isogeny formulae in a given degree $N$, which by-passes the need for factoring division polynomials over large function fields. Using this method, we are able to push the range for which we have formulae at our disposal from $N \leq 13$ to $N \leq 37$ (where in the range $18 \leq N \leq 37$ we have restricted our attention to prime powers). Secondly, using a combination of known techniques and ad-hoc manipulations, we derive optimized versions of these formulae for $N \leq 19$, with some instances performing more than twice as fast as their counterparts from 2020. Thirdly, we solve the problem of understanding the correct choice of radical when walking along the surface between supersingular elliptic curves over $\mathbb{F}_p$ with $p \equiv 7 \bmod 8$; this is non-trivial for even $N$ and was settled for $N = 2$ and $N = 4$ only, in the latter case by Onuki and Moriya at PKC 2022. We give a conjectural statement for all even $N$ and prove it for $N \leq 14$. The speed-ups obtained from these techniques are substantial: using $16$-isogenies, the computation of long chains of $2$-isogenies over $512$-bit prime fields can be accelerated by a factor $3$, and the previous implementation of CSIDH using radical isogenies can be sped up by about $12\%$.

We define the security notion of (strong) collision resistance for chameleon hash functions in the multi-user setting ((S-)MU-CR security). We also present three constructions, CHF_dl, CHF_rsa and CHF_fac, and prove their tight S-MU-CR security based on the discrete logarithm, RSA and factoring assumptions, respectively. In applications, our tightly S-MU-CR secure chameleon hash functions help us to lift a signature scheme from (weak) unforgeability to strong unforgeability in the multi-user setting, and the security reduction is tightness preserving. Furthermore, they can also be used to construct tightly secure online/offline signatures, chameleon signatures and proxy signatures, etc., in the multi-user setting.