International Association
for Cryptologic Research

Descryption. The CWI Cryptology group in Amsterdam and the Mathematical Institute of Leiden University offer a joint PhD position on the topic of Post-Quantum Cryptanalysis. The goal is to advance the state of the art in post-quantum cryptanalysis for the schemes that are currently being standardized. This ranges from improving our understanding in the fundamental computational problems underlying these important quantum-safe schemes, to improving the state of the art in cryptanalytic attacks, e.g., in more refined memory models.

Requirements. Candidates are required to have a master’s degree in mathematics or in computer science. Experience in one or more of these relevant background areas is an advantage: cryptography, algorithms, number theory, coding theory, and quantum algorithms. Some programming skills are expected. Candidates are expected to have an excellent command of English.

Information and application. All applications should include a detailed resume and motivation letter. The application deadline is 31 March 2023. Please visit the vacancy page (click the title) for more information about our terms of employment.

Closing date for applications:

Contact: Marc Stevens (stevens@cwi.nl), Peter Bruin (p.j.bruin@math.leidenuniv.nl)

More information: https://www.cwi.nl/en/jobs/vacancies/983536/

In [15], Leonardi and Ruiz-Lopez propose an additively homomorphic public key encryption scheme whose security is expected to depend on the hardness of the $\textit{learning homomorphism with noise problem}$ (LHN). Choosing parameters for their primitive requires choosing three groups $G$, $H$, and $K$. In their paper, Leonardi and Ruiz-Lopez claim that, when $G$, $H$, and $K$ are abelian, then their public-key cryptosystem is not quantum secure. In this paper, we study security for finite abelian groups $G$, $H$, and $K$ in the classical case. Moreover, we study quantum attacks on instantiations with solvable groups.

We present Fusion, a post-quantum one-time digital signature scheme with non-interactive aggregation with security resting on the short integer solution problem over ideal lattices. Fusion is structurally similar to CRYSTALS-Dilithium, but Fusion is based upon the aggregatable one-time lattice-based scheme by Boneh and Kim. Fusion parameters conservatively target at least 128 bits of security against forgery, taking tightness gaps into account, and with tighter bounds than the BK scheme. Aggregate Fusion signatures are logarithmically sized in the number of keys, so aggregating enough signatures can be more efficient than stacking Dilithium or Falcon signatures.

Guo and Johansson (ASIACRYPT 2021), and MATZOV (tech.~report 2022) have independently claimed improved attacks against various NIST lattice candidate by adding a Fast Fourier Transform (FFT) trick on top of the so-called Dual-Sieve attack. Recently, there was more follow up work in this line adding new practical improvements.

However, from a theoretical perspective, all of these works are painfully specific to Learning with Errors, while the principle of the Dual-Sieve attack is more general (Laarhoven & Walter, CT-RSA 2021). More critically, all of these works are based on heuristics that have received very little theoretical and experimental attention.

This work attempts to rectify the above deficiencies of the literature. We first propose a generalization of the FFT trick by Guo and Johansson to arbitrary Bounded Distance Decoding instances. This generalization offers a new improvement to the attack.

We then theoretically explore the underlying heuristics and show that these are in contradiction with formal, unconditional theorems in some regimes, and with well-tested heuristics in other regimes. The specific instantiations of the recent literature fall into this second regime.

We confirm these contradictions with experiments, documenting several phenomena that are not predicted by the analysis, including a ``waterfall-floor'' phenomenon, reminiscent of Low-Density Parity-Check decoding failures.

We conclude that the success probability of the recent Dual-Sieve-FFT attacks are presumably significantly overestimated. We further discuss the adequate way forward towards fixing the attack and its analysis.

Homomorphic encryption for approximate arithmetic allows one to encrypt discretized real/complex numbers and evaluate arithmetic circuits over them. The first scheme, called CKKS, was introduced by Cheon et al. (Asiacrypt 2017) and gained tremendous attention. The hype for CKKS-type encryption stems from its potential to be used in inference or multiparty computation tasks that do not require the exact output, for example, inference and training of machine learning models.

A desirable property for homomorphic encryption is circuit privacy, which requires that a ciphertext leaks no information on the computation performed to obtain it. Despite numerous improvements, directed toward improving efficiency, the question of circuit privacy for approximate homomorphic encryption remains open.

In this paper, we give the first formal study of circuit privacy for homomorphic encryption over approximate arithmetic. We introduce formal models that allow us to reason about circuit privacy. Then, we show that approximate homomorphic encryption can be made circuit private using tools from differential privacy with appropriately chosen parameters. In particular, we show that by applying an exponential (in the security parameter) Gaussian noise on the evaluated ciphertext, we remove useful information on the circuit from the ciphertext. Crucially, we show that the noise parameter is tight, and taking a lower one leads to an efficient adversary against such a system.

We expand our definitions and analysis to the case of multikey and threshold homomorphic encryption for approximate arithmetic. Such schemes allow users to evaluate a function on their combined inputs and learn the output without leaking anything on the inputs. A special case of multikey and threshold encryption schemes defines a so-called partial decryption algorithm where each user publishes a ``masked'' version of its secret key, allowing all users to decrypt a ciphertext. Similarly, in this case, we show that applying a proper differentially private mechanism gives us IND-CPA-style security where the adversary additionally gets as input the partial decryptions. This is the first security analysis of approximate homomorphic encryption schemes that consider the knowledge of partial decryptions. We show lower bounds on the differential privacy noise that needs to be applied to retain security. Analogously, in the case of circuit privacy, the noise must be exponential in the security parameter. We conclude by showing the impact of the noise on the precision of CKKS-type schemes.

In recent years, the automatic search has been widely used to search differential characteristics and linear approximations with high probability/correlation. Among these methods, the automatic search with the Boolean Satisfiability Problem (SAT, in short) gradually becomes a powerful cryptanalysis tool in symmetric ciphers. A key problem in the automatic search method is how to fully characterize a set $S \subseteq \{0,1\}^n$ with as few Conjunctive Normal Form (CNF, in short) clauses as possible, which is called a full CNF characterization (FCC, in short) problem. In this work, we establish a complete theory to solve a best solution of a FCC problem. Specifically, we start from plain sets, which can be characterized by exactly one clause. By exploring mergeable sets, we find a sufficient and necessary condition for characterizing a plain set. Based on the properties of plain sets, we further provide an algorithm for solving a FCC problem with $S$, which can generate all the minimal plain closures of $S$ and produce a best FCC theoretically. We also apply our algorithm to S-boxes used in block ciphers to characterize their differential distribution tables and linear approximation tables. All of our FCCs are the best-known results at present.

The signed difference is a powerful tool for analyzing the Addition, XOR, Rotation (ARX) cryptographic primitives. Currently, solving the accurate model for the signed difference propagation is infeasible. We propose an approximate MILP modeling method capturing the propagation rules of signed differences. Unlike the accurate signed difference model, the approximate model only focuses on active bits and ignores the possible bit conditions on inactive bits. To overcome the negative effect of a lower accuracy arising from ignoring bit conditions on inactive bits, we propose an additional tool for deducing all bit conditions automatically. Such a tool is based on a directed-graph capturing the whole computation process of ARX primitives by drawing links among intermediate words and operations. The digraph is also applicable in the MILP model construction process: it enables us to identify the parameters upper bounding the number of bit conditions so as to define the objective function; it is further used to connect the boomerang top and bottom signed differential paths by introducing proper constraints to avoid incompatible intersections. Benefiting from the approximate model and the directed-graph based tool, the solving time of the new MILP model is significantly reduced, enabling us to deduce signed differential paths efficiently and accurately.

To show the utility of our method, we propose boomerang attacks on the keyed permutations of three ARX hash functions of BLAKE. For the first time we mount an attack on the full 7 rounds of BLAKE3, with the complexity as low as $2^{180}$. Our best attack on BLAKE2s can improve the previously best result by 0.5 rounds but with lower complexity. The attacks on BLAKE-256 cover the same 8 rounds with the previous best result but with complexity $2^{16}$ times lower. All our results are verified practically with round-reduced boomerang quartets.

We consider a transform, called Derive-then-Derandomize, that hardens a given signature scheme against randomness failure and implementation error. We prove that it works. We then give a general lemma showing indifferentiability of Shrink-MD, a class of constructions that apply a shrinking output transform to an MD-style hash function. Armed with these tools, we give new proofs for the widely standardized and used EdDSA signature scheme, improving prior work in two ways: (1) we give proofs for the case that the hash function is an MD-style one, reflecting the use of SHA512 in the NIST standard, and (2) we improve the tightness of the reduction so that one has guarantees for group sizes in actual use.

This paper introduces protocols for authenticated private information retrieval. These schemes enable a client to fetch a record from a remote database server such that (a) the server does not learn which record the client reads, and (b) the client either obtains the "authentic" record or detects server misbehavior and safely aborts. Both properties are crucial for many applications. Standard private-information-retrieval schemes either do not ensure this form of output authenticity, or they require multiple database replicas with an honest majority. In contrast, we offer multi-server schemes that protect security as long as at least one server is honest. Moreover, if the client can obtain a short digest of the database out of band, then our schemes require only a single server. Performing an authenticated private PGP-public-key lookup on an OpenPGP key server's database of 3.5 million keys (3 GiB), using two non-colluding servers, takes under 1.2 core-seconds of computation, essentially matching the time taken by unauthenticated private information retrieval. Our authenticated single-server schemes are 30-100$\times$ more costly than state-of-the-art unauthenticated single-server schemes, though they achieve incomparably stronger integrity properties.

OpenPubkey makes a client-side modification to OpenID Connect so that an ID Token issued by an OpenID Provider commits to a user held public key. This transforms an ID Token into a certificate that cryptographically binds an OpenID Connect identity to a public key. We call such an ID Token, a PK Token. The user can then sign messages with their signing key and these signatures can be authenticated and attributed to the user’s OpenID Connect identity. This allows OpenPubkey to upgrade OpenID Connect from Bearer Authentication to Proof-of-Possession, eliminating trust assumptions in OpenID Connect and defeating entire categories of attacks present in OpenID Connect. OpenPubkey was designed to satisfy a decade-long need for this functionality. Prior to OpenPubkey, OpenID Connect did not have a secure way for users to sign statements under their OpenID identities.

OpenPubkey is transparent to users and OpenID Providers. An OpenID Provider can not even determine that OpenPubkey is being used. This makes OpenPubkey fully compatible with existing OpenID Providers. In fact a variant of OpenPubkey is currently deployed and used to authenticate signed messages and identities for users with accounts on Google, Microsoft, Okta, and Onelogin. OpenPubkey does not add new trusted parties to OpenID Connect and reduces preexisting trust assumptions. If used in tandem with our MFA-cosigner, OpenPubkey can maintain security even against a malicious OpenID Provider (the most trusted party in OpenID Connect).

An Ideal Cipher (IC) is a cipher where each key defines a random permutation on the domain. Ideal Cipher on a group has many attractive applications, e.g., the Encrypted Key Exchange (EKE) protocol for Password Authenticated Key Exchange (PAKE) [10], or asymmetric PAKE (aPAKE) [40, 36]. However, known constructions for IC on a group domain all have drawbacks, including key leakage from timing information [15], requiring 4 hash-onto-group operations if IC is an 8-round Feistel [27], and limiting the domain to half the group [12] or using variable-time encoding [56, 48] if IC is implemented via (quasi-) bijections from groups to bitstrings [40].

We propose an IC relaxation called a (Randomized) Half-Ideal Cipher (HIC), and we show that HIC on a group can be realized by a modified 2-round Feistel (m2F), at a cost of 1 hash-onto-group operation, which beats existing IC constructions in versatility and computational cost. HIC weakens IC properties by letting part of the ciphertext be non-random, but we exemplify that it can be used as a drop-in replacement for IC by showing that EKE [10] and aPAKE of [40] realize respectively UC PAKE and UC aPAKE even if they use HIC instead of IC. The m2F construction can also serve as IC domain extension, because m2F constructs HIC on domain D from an RO-indiferrentiable hash onto D and an IC on 2κ-bit strings, for κ a security parameter. One application of such extender is a modular lattice-based UC PAKE using EKE instantiated with HIC and anonymous lattice-based KEM.

Whereas theoretical attacks on standardized crypto primitives rarely lead to actual practical attacks, the situation is different for side-channel attacks. Improvements in the performance of side-channel attacks are of utmost importance.

In this paper, we propose a framework to be used in key-recovery side-channel attacks on CCA-secure post-quantum encryption schemes. The basic idea is to construct chosen ciphertext queries to a plaintext checking oracle that collects information on a set of secret variables in a single query. Then a large number of such queries is considered, each related to a different set of secret variables, and they are modeled as a low-density parity-check code (LDPC code). Secret variables are finally determined through efficient iterative decoding methods, such as belief propagation, using soft information. The utilization of LDPC codes offers efficient decoding, source compression, and error correction benefits. It has been demonstrated that this approach provides significant improvements compared to previous work by reducing the required number of queries, such as the number of traces in a power attack.

The framework is demonstrated and implemented in two different cases. On one hand, we attack implementations of HQC in a timing attack, lowering the number of required traces considerably compared to attacks in previous work. On the other hand, we describe and implement a full attack on a masked implementation of Kyber using power analysis. Using the ChipWhisperer evaluation platform, our real-world attacks recover the long-term secret key of a first-order masked implementation of Kyber-768 with an average of only 12 power traces.

The historical domain of information hiding is alternatively used nowadays for communication security. Maybe the oldest and certainly one of the most studied field that falls in this category is steganography. Within the current paper, we focus on image steganography techniques in the case of the JPEG format. We propose a corrected and optimized version of the J3 stegosystem which, to the best of our knowledge and as shown throughout the paper, may be considered a very good choice in terms of security and efficiency as compared to current solutions. We reconstruct the entire J3 algorithm (pre-processing, message insertion and extraction) and detail all the modifications. We also present implementation optimizations and cover all practical cases while still using the maximum image insertion capacity.

We are seeking highly motivated PhDs and post-docs to join the team.

Post-doc positions:
Responsibilities:

Conduct implementation on cutting-edge blockchain technologies and applications.

Develop innovative solutions and proof-of-concepts.

Design blockchain systems.

Collaborate with other team members to ensure successful delivery.
Requirements:

PhD in Computer Science, Electrical Engineering, or related field.

Sufficient knowledge of cryptography and blockchain technology.

Experience with programming languages such as Python, Rust, C++, and Solidity.

Knowledge of distributed systems and peer-to-peer networks.

Strong analytical and problem-solving skills.

Excellent communication and teamwork skills.

PhD positions:
Responsibilities:

Conduct research on cutting-edge cryptography, or AI technologies.

Develop innovative solutions for real-world applications.

Collaborate with national and international academic partners.

Deliver research works to top-tier conferences and journals.
Requirements:

MSc in Computer Science, Math, or related field.

Strong knowledge of AI, or cryptography.

Some experience with programming languages.

Strong analytical and problem-solving skills.

Great communication and teamwork.

Please send your CV, PhD/MSc transcripts, PhD/MSc certificate, English test certificate and a publication list to kaitai.liang@tudelft.nl.
We provide our PhDs and Post-Docs: (1) International academic and industrial collaborations, e.g., working with other top ranking universities, renown companies. (2) Opportunities of participating into various domestic and international cybersecurity projects. (3) Being trained to deliver world-leading research works and publish them to top-tier venues. (4) Flexible and supportive working surroundings. (5) Competitive salary and benefits package, relocation supports, summer and end-year bonus, free academic trainings.

Closing date for applications:

Contact: K. Liang (kaitai.liang@tudelft.nl)

In this work, we present a novel generic construction for a Distributed Key Generation (DKG) scheme. Our generic construction relies on three modular cryptographic building blocks. The first is an aggregatable Verifiable Secret Sharing (AgVSS) scheme, the second is a Non-Interactive Key Exchange (NIKE) scheme, and the third is a secure hash function. We give formal definitions for the AgVSS and NIKE schemes, as well as concrete constructions. The utility of this generic construction is flexibility; i.e., any aggregatable VSS and NIKE scheme can be employed, and the construction will remain secure.

To prove the security of our generic construction, we introduce formalized game-based notions of security for DKGs, building upon existing notions in the literature. However, these prior security notions either were presented informally, omitted important requirements, or assumed certain algebraic structure of the underlying scheme. Our security notions make no such assumption of underlying algebraic structure, and explicitly consider details such as participant consistency, communication patterns, and key validity. Further, our security notions imply simulatability with respect to a target key generation scheme without rewinding. Hence, any construction that is proven secure using our security notions additionally achieves UC security.

We then present STORM, a concrete instantiation of our generic construction that is secure in the discrete logarithm setting in the random oracle model. STORM is more efficient than related DKG schemes in the literature. Because of its simple design and composability, it is a practical choice for real-world settings and standardization efforts.

The growing popularity of cloud storage has brought attention to critical need for preventing information leakage from cloud access patterns. To this end, recent efforts have extended Oblivious RAM (ORAM) to the cloud environment in the form of Oblivious Store. However, its impracticality due to the use of probability encryption with fake accesses to obfuscate the access pattern, as well as the security requirements of conventional obliviousness designs, which hinder cloud interests in improving storage utilization by removing redundant data among cross-users, limit its effectiveness. Thus, we propose a practical Oblivious Store, PEO-Store, which integrates the obliviousness property into the cloud while removing redundancy without compromising security. Unlike conventional schemes, PEO-Store randomly selects a delegate for each client to communicate with the cloud, breaking the mapping link between a valid access pattern sequence and a specific client. Each client encrypts their data and shares it with selected delegates, who act as intermediaries with the cloud provider. This design leverages non-interactive zero-knowledge-based redundancy detection, discrete logarithm problem-based key sharing, and secure time-based delivery proof to protect access pattern privacy and accurately identify and remove redundancy in the cloud. The theoretical proof demonstrates that the probability of identifying the valid access pattern with a specific user is negligible in our design. Experimental results show that PEO-Store outperforms state-of-the-art methods, achieving an average throughput of up to 3 times faster and saving 74% of storage space.

In this short note, we describe a few implementation techniques that allow performing key pair generation for the Falcon and Hawk lattice-based signature schemes, and for the BAT key encapsulation scheme, in a fully constant-time way and without any use of floating-point operations. Our new code is faster than previously published implementations, especially when running on small embedded systems, and uses less RAM.

A secret-sharing scheme enables a dealer, holding a secret string, to distribute shares to parties such that only pre-defined authorized subsets of parties can reconstruct the secret. The collection of authorized sets is called an access structure. There is a huge gap between the best known upper-bounds on the share size of a secret-sharing scheme realizing an arbitrary access structure and the best known lower-bounds on the size of these shares. For an arbitrary $n$-party access structure, the best known upper-bound on the share size is $2^{O(n)}$. On the other hand, the best known lower-bound on the total share size is much smaller, i.e., $\Omega(n^2/\log (n))$ [Csirmaz, \emph{Studia Sci. Math. Hungar.}]. This lower-bound was proved more than 25 years ago and no major progress was made since.

In this paper, we study secret-sharing schemes for k-hypergraphs, i.e., for access structure where all minimal authorized sets are of size exactly $k$ (however, unauthorized sets can be larger). We consider the case where $k$ is small, i.e., constant or at most $\log (n)$. The trivial upper-bound for these access structures is $O(k\cdot \binom{n}{k})$ and this can be slightly improved. If there were efficient secret-sharing schemes for such $k$-hypergraphs (e.g., $2$-hypergraphs or $3$-hypergraphs), then we would be able to construct secret-sharing schemes for arbitrary access structure that are better than the best known schemes. Thus, understanding the share size required for $k$-hypergraphs is important. Prior to our work, the best known lower-bound for these access structures was $\Omega(n \log (n))$, which holds already for graphs (i.e., $2$-hypergraphs).

We improve this lower-bound, proving a lower-bound of $\Omega(n^{1-1/(k-1)}/k)$ for some explicit $k$-hypergraphs, where $3 \leq k \leq \log (n)$. For example, for $3$-hypergraphs we prove a lower-bound of $\Omega(n^{3/2})$. For $\log (n)$-hypergraphs, we prove a lower-bound of $\Omega(n^{2}/\log (n))$, i.e., we show that the lower-bound of Csirmaz holds already when all minimal authorized sets are of size $\log (n)$. Our proof is simple and shows that the lower-bound of Csirmaz holds for a simple variant of the access structure considered by Csirmaz.

A central problem in cryptanalysis is to find all the significant deviations from randomness in a given $n$-bit cryptographic primitive. When $n$ is small (e.g., an $8$-bit S-box), this is easy to do, but for large $n$, the only practical way to find such statistical properties was to exploit the internal structure of the primitive and to speed up the search with a variety of heuristic rules of thumb. However, such bottom-up techniques can miss many properties, especially in cryptosystems which are designed to have hidden trapdoors.

In this paper we consider the top-down version of the problem in which the cryptographic primitive is given as a structureless black box, and reduce the complexity of the best known techniques for finding all its significant differential and linear properties by a large factor of $2^{n/2}$. Our main new tool is the idea of using {\it surrogate differentiation}. In the context of finding differential properties, it enables us to simultaneously find information about all the differentials of the form $f(x) \oplus f(x \oplus \alpha)$ in all possible directions $\alpha$ by differentiating $f$ in a single arbitrarily chosen direction $\gamma$ (which is unrelated to the $\alpha$'s). In the context of finding linear properties, surrogate differentiation can be combined in a highly effective way with the Fast Fourier Transform. For $64$-bit cryptographic primitives, this technique makes it possible to automatically find in about $2^{64}$ time all their differentials with probability $p \geq 2^{-32}$ and all their linear approximations with bias $|p| \geq 2^{-16}$; previous algorithms for these problems required at least $2^{96}$ time. Similar techniques can be used to significantly improve the best known time complexities of finding related key differentials, second-order differentials, and boomerangs. In addition, we show how to run variants of these algorithms which require no memory, and how to detect such statistical properties even in trapdoored cryptosystems whose designers specifically try to evade our techniques.

Recent advancements in low-cost cryptography have converged upon the use of nanoscale level structural variances as sources of entropy that is unique to each device. Consequently, such delay-based Physically Unclonable Functions or (PUFs) have gained traction for several cryptographic applications. In light of recent machine learning (ML) attacks on delay-based PUFs, the common trend among PUF designers is to either introduce non-linearity using XORs or input transformations applied on the challenges in order to harden the security of delay-based PUFs. Such approaches make machine learning modelling attacks hard by destroying the linear relationship between challenge-response pairs of a PUF. However, we propose to perceive PUFs, which are fundamentally viewed as Boolean functional mapping, as a set of delay parameters drawn from normal distribution. Using this newfound perception, we propose an alternative attack strategy in this paper. We show that instead of trying to learn the exact functional relationship between challenge-response pairs from a PUF, one can search through the search space of all PUFs to find alternative PUF delay parameter set that exhibits similar behaviour as the target PUF. The core intuition behind this strategy is that one can consider a PUF as a set of stages wherein, depending on the corresponding input challenge bit, one of the several signals within a PUF's stage win a race condition. To utilize this idea, we develop a novel Particle Swarm Optimization based framework inspired by the biomimicry of amoebic reproduction. The proposed algorithm avoids the pitfalls of textbook Genetic Algorithms and demonstrates complete break of existing delay-based PUFs which are based on arbiter chains. More specifically, we are able to model higher-rder $k$-XOR PUF variants which are resistant to all-known ML modelling techniques, including $k=13, 15$ and $20$, without the knowledge of reliability values. In addition to that, we also model PUFs that incorporate input transformation, like variants of IPUF and LP-PUF. Furthermore, we take forward this idea across different search spaces in order to learn a higher order PUF using a lower order (and simpler) PUF architecture. This allows us to explore a novel class of attacks, including modelling a $k$-XOR PUF using a $(k-1)$-XOR PUF as well as bypassing input transformations based PUF designs.