International Association
for Cryptologic Research

Zero-knowledge proof (ZKP) is a cryptographic primitive that enables one party to prove the validity of a statement to other parties without disclosing any secret information. With its widespread adoption in applications such as blockchain and verifiable machine learning, the demand for generating zero-knowledge proofs has increased dramatically. In recent years, considerable efforts have been directed toward developing GPU-accelerated systems for proof generation. However, these previous systems only explored efficiently generating a single proof by reducing latency rather than batch generation to provide high throughput.

We propose a fully pipelined GPU-accelerated system for batch generation of zero-knowledge proofs. Our system has three features to improve throughput. First, we design a pipelined approach that enables each GPU thread to continuously execute its designated proof generation task without being idle. Second, our system supports recent efficient ZKP protocols with their computational modules: sum-check protocol, Merkle tree, and linear-time encoder. We customize these modules to fit our pipelined execution. Third, we adopt a dynamic loading method for the data required for proof generation, reducing the required device memory. Moreover, multi-stream technology enables the overlap of data transfers and GPU computations, reducing overhead caused by data exchanges between host and device memory.

We implement our system and evaluate it on various GPU cards. The results show that our system achieves more than 259.5× higher throughput compared to state-of-the-art GPU-accelerated systems. Moreover, we deploy our system in the verifiable machine learning application, where our system generates 9.52 proofs per second, successfully achieving sub-second proof generation for the first time in this field.

Let $N=pq$ be the product of two balanced prime numbers $p$ and $q$. In 2015, Roman'kov introduced an interesting RSA-like cryptosystem that, unlike the classical RSA key equation $ed - k (p-1)(q-1) = 1$, uses the key equation $ed - k r = 1$, where $r | p-1$ and is a large prime number. In this paper, we study if small private key attacks based on lattices can be applied to Roman'kov's cryptosystem. More precisely, we argue that such attacks do not appear to be applicable to this scheme.

The construction of self-orthogonal codes from functions over finite fields has been widely studied in the literature. In this paper, we construct new families of self-orthogonal linear codes with few weights from trace functions and weakly regular plateaued functions over the finite fields of odd characteristics. We determine all parameters of the constructed self-orthogonal codes and their dual codes. Moreover, we employ the constructed $p$-ary self-orthogonal codes to construct $p$-ary LCD codes.

As privacy concerns have arisen in machine learning, privacy-preserving machine learning (PPML) has received significant attention. Fully homomorphic encryption (FHE) and secure multi-party computation (MPC) are representative building blocks for PPML. However, in PPML protocols based on FHE and MPC, interaction between the client (who provides encrypted input data) and the evaluator (who performs the computation) is essential to obtain the final result in plaintext. Functional encryption (FE) is a promising candidate to remove this constraint, but existing FE-based PPML protocols are restricted to evaluating only simple ML models, such as one-layer neural networks, or they support partially encrypted PPML, which makes them vulnerable to information leakage beyond the inference results.

In this paper, we propose a fully encrypted FE-based PPML protocol, which supports the evaluation of arbitrary functions over encrypted data with no information leakage during computation, for the first time. To achieve this, we newly construct a vector functional encryption scheme for quadratic polynomials and combine it with an inner product encryption scheme. This enables multiple compositions of quadratic polynomials to compute arbitrary complex functions in an encrypted manner.

Our FE-based PPML protocol is secure in the malicious model, which means that an adversary cannot obtain any information about the input data even though they intentionally deviate from the protocol. We then show how to use our protocol to build a fully encrypted 2-layer neural network model with quadratic activation functions and present experimental results.

Boneh et al. (CRYPTO'18) proposed two $t$-out-of-$N$ threshold fully homomorphic encryption ($\sf TFHE$) schemes based on Shamir secret sharing scheme and $\{0,1\}$-linear secret sharing scheme. They demonstrated the simulation security, ensuring no information leakage during partial or final decryption. This breakthrough allows any scheme to be converted into a threshold scheme by using $\sf TFHE$.

We propose two polynomial time algorithms to break the simulation security of $t$-out-of-$N$ $\sf TFHE$ based on Shamir secret sharing scheme proposed by Boneh et al.. First, we show that an adversary can break the simulation security by recovering the secret key under some constraints on $t$ and $N$, which does not violate the conditions for security proof. Next, we introduce a straightforward fix that theoretically satisfies the simulation security. However, we argue that this modification remains insecure insecure when implemented with any state-of-the-art fully homomorphic encryption libraries in practice. To ensure robustness against our subsequent attacks, we recommend using an error-refreshing algorithm, such as bootstrapping or modulus switching, for each addition operation.

We extend the concept of access control for functional encryption, introduced by Abdalla et al. (ASIACRYPT 2020), to function-revealing encryption (Joy and Passelègue, SCN 2018). Here “access control” means that function evaluation is only possible when a specified access policy is met. Specifically, we introduce access-controlled inner-product function-revealing encryption (AC-IPFRE) and give two applications.

On the theoretical side, we use AC-IPFRE to show that function- hiding inner-product functional encryption (FH-IPFE), introduced by Bishop et al. (ASIACRYPT 2015), is equivalent to IPFRE. To show this, we in particular generically construct AC-IPFRE from IPFRE for the “non-zero inner-product” (NZIP) access policy. This result uses an effective version of Lagrange’s Four Square Theorem. One consequence of this result is that lower bounds by Ünal (EUROCRYPT 2020) suggest that, as for FH-IPFE, bilinear pairings will be needed to build IPFRE.

On the practical side, we build an outsourced approximate nearest- neighbor (ANN) search protocol and mitigate its leakage via AC-IPFRE. For this, we construct a practical AC-IPFRE scheme in the generic bilinear group model for a specific access policy for ANN search. To this end, we show that techniques of Wee (TCC 2020) implicitly give the most practical FH-IPFE scheme to date. We implement the resulting outsourced ANN search protocol and report on its performance.

Of independent interest, we show AC-IPFRE for NZIP implies attribute-hiding small-universe AC-IPFRE for arbitrary access policies. Previous work on access control for FE did not achieve attribute hiding. Overall, our results demonstrate that AC-IPFRE is of both theoretical and practical interest and set the stage for future work in the area.

Modern micro-architectural attacks use a variety of building blocks chained to develop a final exploit. However, since in most cases, the footprint of such attacks is not visible architecturally (like, in the file-system), it becomes trickier to defend against these. In light of this, several automated defence mechanisms use Hardware Performance Counters (HPCs) detect when the micro-architectural elements are being misused for a potential attacks (like flush-reload, Spectre, Meltdown etc.). In order to bypass such defences, recent works have proposed the idea of "probabilistic interleaving": the adversary interleaves the actual attack code with benign code with very low frequency. Such a strategy tips off the HPCs used for detection with a lot of unnecessary noise; recent studies have shown that probabilistically interleaved attacks can achieve an attack evasion rate of 100% (i.e. are virtually undetectable).

In this work, we contend this folklore. We develop a theoretical model of interleaved attacks using lightweight statistical tools like Gaussian Mixture Models and Dip Test for Unimodality and prove they are detectable for the correct choices of HPCs. Furthermore, we also show possible defence strategy against a stronger threat model than considered in literature: where the attacker interleaves multiple attacks instead of a single attack. Empirically, to instantiate our detector, in contrast to prior detection strategies, we choose LLMs for a number of reasons: (1) LLMs can easily contextualize data from a larger set of HPCs than generic machine learning techniques, and (2) with simple prompts, LLMs can quickly switch between different statistical analysis methods. To this end, we develop an LLM-based methodology to detect probabilistically interleaved attacks. Our experiments establish that our improved methodology is able to achieve 100% speculative attacks like Spectre v1/v2/v3, Meltdown, and Spectre v2 (with improved gadgets that even evade recent protections like Enhanced IBRS, IBPB conditional, and so on). This makes our methodology suitable for detecting speculative attacks in a non-profiled setting: where attack signatures might not be known in advance. All in all, we achieve a 100% attack detection rate, even with very low interleave frequencies (i.e. $10^{-6}$). Our detection principle and its instantiation through LLMs shows how probabilistically interleaving attack code in benign execution is not a perfect strategy, and more research is still needed into developing and countering better attack evasion strategies.

Efficiently verifying mathematical proofs and computations has been a heavily researched topic within Computer Science. Particularly, even repetitive steps within a proof become much more complex and inefficient to validate as proof sizes grow. To solve this problem, we suggest viewing it through the lens of Incrementally Verifiable Computation (IVC). However, many IVC methods, including the state-of-the-art Nova recursive SNARKs, require proofs to be linear and for each proof step to be identical. This paper proposes Lova, a novel framework to verify mathematical proofs end-to-end that solves these problems. Particularly, our approach achieves a few novelties alongside the first-of-its-kind implementation of Nova: (i) an innovative proof splicing mechanism to generate independent proof sequences, (ii) a system of linear algorithms to verify a variety of mathematical logic rules, and (iii) a novel multiplexing circuit allowing non-homogeneous proof sequences to be verified together in a single Nova proof. The resulting Lova pipeline has linear prover time, constant verifying capability, dynamic/easy modification, and optional zero-knowledge privacy to efficiently validate mathematical proofs. Code is available at https://github.com/noelkelias/lova.

We show that for every polynomial q∗ there exist polynomial-size, constant-query, non-adaptive PCPs for NP which are perfect zero knowledge against (adaptive) adversaries making at most q∗ queries to the proof. In addition, we construct exponential-size constant-query PCPs for NEXP with perfect zero knowledge against any polynomial-time adversary. This improves upon both a recent construction of perfect zero-knowledge PCPs for #P (STOC 2024) and the seminal work of Kilian, Petrank and Tardos (STOC 1997).

NTRU-like constructions are among the most studied lattice-based schemes. The freedom of design of NTRU resulted in many variants in literature motivated by faster computations or more resistance against lattice attacks by changing the underlying algebra. To the best of our knowledge, BQTRU (DCC 2017), a noncommutative NTRU-like cryptosystem, is the fastest claimed variant of NTRU built over the quaternion algebra of the bivariate ring of polynomials. The key generation and the encryption of BQTRU are claimed to be 16/7 times faster than standard NTRU for equivalent levels of security. For key recovery attacks, the authors claim that retrieving a decryption key is equivalent to solving the Shortest Vector Problem (SVP) in expanded Euclidean lattices of giant dimensions. This work disproves this claim and proposes practical key and message recovery attacks that break the moderate parameter sets of BQTRU estimated to achieve $2^{92}$ message security and $2^{166}$ key security on a standard desktop within less than two core weeks. Furthermore, our analysis shows that the proposed parameter set for the highest security level claiming $2^{212}$ message security and $2^{396}$ key security can barely achieve $2^{82}$ message security and $2^{125}$ key security. Our work not only provides cryptanalysis for BQTRU but also demonstrates the potential of extending Gentry's attack to other rings beyond the cyclotomic polynomial ring.

Any isogeny between two supersingular elliptic curves can be defined over $\mathbb{F}_{p^2}$, however, this does not imply that computing such isogenies can be done with field operations in $\mathbb{F}_{p^2}$. In fact, the kernel generators of such isogenies are defined over extension fields of $\mathbb{F}_{p^2}$, generically with extension degree linear to the isogeny degree. Most algorithms related to isogeny computations are only efficient when the extension degree is small. This leads to efficient algorithms used in isogeny-based cryptographic constructions, but also limits their parameter choices at the same time. In this paper, we consider three computational subroutines regarding isogenies, focusing on cases with large extension degrees: computing a basis of $\ell$-torsion points, computing the kernel polynomial of an isogeny given a kernel generator, and computing the kernel generator of an isogeny given the corresponding quaternion ideal under the Deuring correspondence. We then apply our algorithms to the constructive Deuring correspondence algorithm from Eriksen, Panny, Sotáková and Veroni (LuCaNT'23) in the case of a generic prime characteristic, achieving around 30% speedup over their results.

Protein sequence classification is crucial in many research areas, such as predicting protein structures and discovering new protein functions. Leveraging large language models (LLMs) is greatly promising to enhance our ability to tackle protein sequence classification problems; however, the accompanying privacy issues are becoming increasingly prominent. In this paper, we present a privacy-preserving, non-interactive, efficient, and accurate protocol called encrypted DASHformer to evaluate a transformer-based neural network for protein sequence classification named DASHformer, provided by the iDASH 2024-Track 1 competition. The presented protocol is based on our solution for this competition, which won the first place. It is arguably the first secure transformer inference protocol capable of performing batch classification for multiple protein sequences in a single execution only using leveled homomorphic encryption (i.e., without bootstrapping). To achieve this, we propose a series of new techniques and algorithmic improvements, including data-driven non-polynomial function fitting, tensor packing, and double baby-step-giant-step for computing the product of multiple encrypted matrices. These techniques and improvements enable the protocol to classify $163$ encrypted protein sequences in about $165$ seconds with $128$-bit security, achieving an amortized time of about one second per sequence.

In response to the quantum threat, new post-quantum cryptographic algorithms will soon be deployed to replace existing public-key schemes. MAYO is a quantum-resistant digital signature scheme whose small keys and signatures make it suitable for widespread adoption, including on embedded platforms with limited security resources. This paper demonstrates two single-trace side-channel attacks on a MAYO implementation in ARM Cortex-M4 that recover a secret key with probabilities of 99.9% and 91.6%, respectively. Both attacks use deep learning-assisted power analysis exploiting information leakage during modular multiplication to reveal a vector in the oil space. This vector is then extended to a full secret key using algebraic techniques.

The analytical perspective employed in the study classifies the theoretical research on dependencies in differential characteristics into two types. By categorising all dependence representations from the value restrictions and the theory of quasidifferential trails, we pinpoint a specific set of nonlinear constraints, which we term linearised nonlinear constraints. We aim to establish a method that utilises value restrictions to identify these constraints, as the current method based on value restrictions is found to be lacking in this area. A linearisation method for searching linearised nonlinear constraints for a given differential characteristic is developed by leveraging linear dependencies between inputs and outputs of active S-boxes. Then, we propose a three-stage evaluation approach to more accurately evaluate differential characteristics with linearised nonlinear constraints. Four differential characteristics of GIFT-64 are analysed using the three-stage evaluation approach, and the exact right key spaces and remaining probabilities are given. According to our results, the right key spaces of the four differential characteristics do not cover the entire key space, and the remaining probabilities are not equivalent to the stated probabilities. Concerning GIFT-128, we find six differential characteristics subject to linearised nonlinear constraints. Besides, inconsistencies are detected in the linear and linearised nonlinear constraints in the characteristics of two differentials employed to initiate the most effective differential attack on GIFT-128. Based on these results, we strongly advise reassessing the differential attacks that rely on these distinguishers. An additional advantage of using the linearisation method and the three-stage evaluation approach is their ability to identify linear and nonlinear constraints in ciphers that utilise the Generalised Feistel Network (GFN). It leads to the first instantiations of linear and nonlinear constraints in the GFN cipher WARP.

We're on a mission to solve the foundational scientific and engineering problems that stand in the way of mainstream blockchain adoption. We're starting by focusing on the areas of interoperability, scalability, and usability. But we can't conquer this frontier alone, we're therefore on the hunt for brilliant minds to join us. We're primarily seeking senior engineers to join our team, though we have exciting opportunities across other roles as well. View our open positions at https://commonprefix.com/careers

Closing date for applications:

Contact: Dimitris Lamprinos ([email protected])

More information: https://commonprefix.com

We seek to hire a full-time postdoctoral researcher in the area of security and privacy.

The University offers:

• A 2.5-year contract at an exciting international environment.
• Generous travel funds.
• Possibility to co-supervise PhD students.

Your role:

The successful candidate is expected to contribute to the PROVTOPIA project, which focuses on counteracting disinformation via Secure and Private Provenance Verification of Media Content. The candidate will work under the umbrella of the Crises research group (https://crises-deim.urv.cat/) and the direction of Dr. Rolando Trujillo. Candidates with experience in applied cryptography, threat modelling or formal verification are encouraged to apply.

Include in your application the following documents:

• Curriculum Vitae
• Research statement
• Contact information for 3 referees

Deadline for applications is 15 January 2025 . Early applications are highly encouraged, though, as they will be processed upon reception.

Closing date for applications:

Contact: Dr. Rolando Trujillo ([email protected])

More information: https://rolandotr.bitbucket.io/open-positions.html

Aalto University School of Electrical Engineering is hiring new faculty members. The ideal candidate will have expertise in fields, such as but not limited to, hardware-accelerated computing, heterogeneous architectures, compiler technologies and code generation, or the co-design of software and hardware systems. The application closes on 29.12.2024. For more information, see https://www.aalto.fi/en/open-positions/professor-computer-engineering.

Closing date for applications:

Contact: For additional information about the position, please contact Professor Riku Jäntti at riku.jantti at aalto.fi. For questions related to the recruitment process, HR Partner Hanna Koli at hanna.koli at aalto.fi.

More information: https://www.aalto.fi/en/open-positions/professor-computer-engineering

We introduce the notion of non-interactive zero-knowledge (NIZK) proofs with certified deletion. Our notion enables the recipient of a quantum NIZK proof for a (quantumly hard) NP statement to delete the proof and collapse it into a classical deletion certificate. Once this certificate is successfully validated, we require the recipient of the proof to lose their ability to find accepting inputs to NIZK verification.

We formally define this notion and build several candidate constructions from standard cryptographic assumptions. In particular, we propose a primary construction from classical NIZK for NP and one-way functions, albeit with two limitations: (i) deletion certificates are only privately verifiable, and (ii) both prover and verifier are required to be quantum algorithms. We resolve these hurdles in two extensions that assume the quantum hardness of the learning with errors problem. The first one achieves publicly verifiable certificates, and the second one requires merely classical communication between classical provers and quantum verifiers.

Non-Interactive Zero-Knowledge Arguments (NIZKs) are cryptographic protocols that enable a prover to demonstrate the validity of an $\mathsf{NP}$ statement to a verifier with a single message, without revealing any additional information. The soundness and zero-knowledge properties of a NIZK correspond to security against a malicious prover and a malicious verifier respectively. Statistical NIZKs (S-NIZKs) are a variant of NIZKs for which the zero-knowledge property is guaranteed to hold information-theoretically. Previous works have shown that S-NIZKs satisfying a weak version of soundness known as static soundness exist based on standard assumptions. However, the work of Pass (TCC 2013) showed that S-NIZKs with the stronger \emph{adaptive} soundness property are inherently challenging to obtain. The work proved that standard (black-box) proof techniques are insufficient to prove the security of an S-NIZK based on any standard (falsifiable) assumption. We extend this result to the setting where parties can perform quantum computations and communicate using quantum information, while the quantum security reduction is restricted to query the adversary classically. To this end, we adapt the well-known meta-reduction paradigm for showing impossibility results to the quantum setting. Additionally, we reinterpret our result using a new framework for studying quantum reductions, which we believe to be of independent interest.

The hardness of lattice problems offers one of the most promising security foundations for quantum-safe cryptography. Basic schemes for public key encryption and digital signatures are already close to standardization at NIST and several other standardization bodies, and the research frontier has moved on to building primitives with more advanced privacy features. At the core of many such primi- tives are zero-knowledge proofs. In recent years, zero-knowledge proofs for (and using) lattice relations have seen a dramatic jump in efficiency and they currently provide arguably the shortest, and most computationally efficient, quantum-safe proofs for many sce- narios. The main difficulty in using these proofs by non-experts (and experts!) is that they have a lot of moving parts and a lot of internal parameters depend on the particular instance that one is trying to prove. Our main contribution is a library for zero-knowledge and suc- cinct proofs which consists of efficient and flexible C code under- neath a simple-to-use Python interface. Users without any back- ground in lattice-based proofs should be able to specify the lattice relations and the norm bounds that they would like to prove and the library will automatically create a proof system, complete with the intrinsic parameters, using either the succinct proofs of LaBRADOR (Beullens and Seiler, Crypto 2023) or the linear-size, though smaller for certain application, proofs of Lyubashevsky et al. (Crypto 2022). The Python interface also allows for common operations used in lattice-based cryptography which will enable users to write and pro- totype their full protocols within the syntactically simple Python environment. We showcase some of the library’s usefulness by giving protocol implementations for blind signatures, anonymous credentials, the zero-knowledge proof needed in the recent Swoosh protocol (Gaj- land et al., Usenix 2024), proving knowledge of Kyber keys, and an aggregate signature scheme. Most of these are the most efficient, from a size, speed, and memory perspective, known quantum-safe instantiations.