International Association
for Cryptologic Research

Event date: 17 August to 21 August 2025
Submission deadline: 13 February 2025
Notification: 3 May 2025

The Messaging Layer Security (MLS) protocol, recently standardized in RFC 9420, aims to provide efficient asynchronous group key establishment with strong security guarantees. The main component of MLS, which is the source of its important efficiency and security properties, is a protocol called TreeKEM. Given that a major vision for the MLS protocol is for it to become the new standard for messaging applications like WhatsApp, Facebook Messenger, Signal, etc., it has the potential to be used by a huge number of users. Thus, it is important to better understand the security of MLS and hence also of TreeKEM. In a previous work by Klein et. al, TreeKEM was proven adaptively secure in the Random Oracle Model (ROM) with a polynomial loss in security by proving a result about the security of an arbitrary IND-CPA secure public-key encryption scheme in a public-key version of the Generalized Selective Decryption (GSD) security game.

In this work, we prove a tighter bound for the security of TreeKEM. We follow the approach in the aforementioned work and first introduce a modified version of the public-key GSD game better suited for analyzing TreeKEM. We then provide a simple and detailed proof of security for a specific encryption scheme, the DHIES scheme (currently the only standardized scheme in MLS), in this game in the ROM and achieve a tighter bound compared to the result from Klein et. al. We also define and describe the syntax and security of TreeKEM-like schemes and state a result linking the security of TreeKEM with security in our GSD game in the ROM.

We initiate the study of the black-box complexity of private-key functional encryption (FE). Of central importance in the private-key setting is the inner-product functionality, which is currently only known from assumptions that imply public-key encryption, such as Decisional Diffie-Hellman or Learning-with-Errors. As our main result, we rule out black-box constructions of private-key inner-product FE from random oracles. This implies a black-box separation between private-key inner-product FE from all symmetric-key primitives implied by random oracles (e.g., symmetric-key encryption and collision-resistant hash functions).

Proving lower bounds for private-key functional encryption schemes introduces challenges that were absent in prior works. In particular, the combinatorial techniques developed by prior works for proving black-box lower bounds are only useful in the public-key setting and predicate encryption settings, which all fail for the private-key FE case. Our work develops novel combinatorial techniques based on Fourier analysis to overcome these barriers. We expect these techniques to be widely useful in future research in this area.

Given the devastating security compromises caused by side-channel attacks on existing classical systems, can we store our private data encoded as a quantum state so that they can be kept private in the face of arbitrary side-channel attacks?

The unclonable nature of quantum information allows us to build various quantum protection schemes for cryptographic information such as secret keys. Examples of quantum protection notions include copy-protection, secure leasing, and finally, unbounded leakage-resilience, which was recently introduced by Çakan, Goyal, Liu-Zhang and Ribeiro (TCC'24). Çakan et al show that secrets of various cryptographic schemes (such as cryptographic keys or secret shares) can be protected by storing them as quantum states so that they satisfy LOCC (local operation and classical communication) leakage-resilience: the scheme can tolerate any unbounded amount of adaptive leakage over unbounded rounds. As a special case (dubbed $1$-round leakage), this also means that those quantum states cannot be converted to classical strings (without completely losing their functionality).

In this work, we continue the study of unbounded/LOCC leakage-resilience and consider several new primitive. In more details, we build ciphertexts, signatures and non-interactive zero-knowledge proofs with unbounded leakage-resilience. We show the following results.

- Assuming the existence of a classical $X \in \{\text{secret-key encryption}, \text{public-key encryption}\}$ scheme, we construct an $X$ scheme with LOCC leakage-resilient ciphertexts. This guarantees that an adversary who obtains LOCC-leakage on ciphertexts cannot learn anything about their contents, even if they obtain the secret key later on.

- Assuming the existence of a classical signature scheme and indistinguishability obfuscation (iO), we construct a signature scheme with LOCC leakage-resilient signatures. This guarantees that an adversary who obtains LOCC-leakage on various signatures cannot produce any valid signatures at all other than the ones it obtained honestly!

- Assuming the existence of one-way functions and indistinguishability obfuscation (iO), we construct a NIZK proof system with LOCC leakage-resilient proofs. This guarantees that an adversary who obtains LOCC-leakage on a NIZK proof of an hard instance cannot produce a valid proof!

The National Institute for Standards and Technology (NIST) initiated a standardization procedure for additional digital signatures and recently announced round-2 candidates for the PQ additional digital signature schemes. The multivariate digital signature scheme Unbalanced Oil and Vinegar (UOV) is one of the oldest post-quantum schemes and has been selected by NIST for Round 2. Although UOV is mathematically secure, several side-channel attacks (SCA) have been shown on the UOV or UOV-based digital signatures. We carefully analyze the sensitivity of variables and operations in the UOV scheme from the side-channel perspective and show which require protection. To mitigate implementation-based SCA, we integrate a provably secure arbitrary-order masking technique with the key generation and signature generation algorithms of UOV. We propose efficient techniques for the masked dot-product and matrix-vector operations, which are both critical in multivariate DS schemes. We also implemented and demonstrate the practical feasibility of our masking algorithms for UOV-Ip on the ARM Cortex-M4 microcontroller. Our first-order masked UOV implementations have $2.7\times$ and $3.6\times$ performance overhead compared to the unmasked scheme for key generation and signature generation algorithms. Our first-order masked UOV implementations use $1.3\times$ and $1.9\times$ stack memory rather than the unmasked version of the key generation and signature generation algorithms.

The eIDAS 2.0 regulation aims to develop interoperable digital identities for European citizens, and it has recently become law. One of its requirements is that credentials be unlinkable. Anonymous credentials (AC) allow holders to prove statements about their identity in a way that does not require to reveal their identity and does not enable linking different usages of the same credential. As a result, they are likely to become the technology that provides digital identity for Europeans.

Any digital credential system, including anonymous credentials, needs to be secured against identity theft and fraud. In this work, we introduce the notion of a multi-holder anonymous credential scheme that allows issuing shares of credentials to different authentication factors (or ``holders''). To present the credential, the user's authentication factors jointly run a threshold presentation protocol. Our definition of security requires that the scheme provide unforgeability: the adversary cannot succeed in presenting a credential with identity attributes that do not correspond to an identity for which the adversary controls at least $t$ shares; this is true even if the adversary can obtain credentials of its choice and cause concurrent executions of the presentation protocol. Further, our definition requires that the presentation protocol provide security with identifiable abort. Finally, presentations generated by all honest holders must be unlinkable and must not reveal the user's secret identity attributes even to an adversary that controls some of the user's authentication factors.

We design and prove the (concurrent) security of a multi-holder version of the BBS anonymous credential scheme. In our construction, each holder is issued a secret share of a BBS credential. Using these shares, the holders jointly compute a credential presentation that is identical to (and therefore compatible with) the traditional, single-holder variant (due to Tessaro and Zhu, Eurocrypt'23) of a BBS credential presentation.

As Succinct Non-interactive Arguments of Knowledge (SNARKs) gain traction for large-scale applications, distributed proof generation is a promising technique to horizontally scale up the performance. In such protocols, the workload to generate SNARK proofs is distributed among a set of workers, potentially with the help of a coordinator. Desiderata include linear worker time (in the size of their sub-tasks), low coordination overhead, low communication complexity, and accountability (the coordinator can identify malicious workers). State-of-the-art schemes, however, do not achieve these properties.

In this paper, we introduced $\mathsf{Cirrus}$, the first accountable distributed proof generation protocol with linear computation complexity for all parties. $\mathsf{Cirrus}$ is based on HyperPlonk (EUROCRYPT'23) and therefore supports a universal trusted setup. $\mathsf{Cirrus}$ is horizontally scalable: proving statements about a circuit of size $O(MT)$ takes $O(T)$ time with $M$ workers. The per-machine communication cost of $\mathsf{Cirrus}$ is low, which is only logarithmic in the size of each sub-circuit. $\mathsf{Cirrus}$ is also accountable, and the verification overhead of the coordinator is efficient. We further devised a load balancing technique to make the workload of the coordinator independent of the size of each sub-circuit.

We implemented an end-to-end prototype of $\mathsf{Cirrus}$ and evaluated its performance on modestly powerful machines. Our results confirm the horizontal scalability of $\mathsf{Cirrus}$, and the proof generation time for circuits with $2^{25}$ gates is roughly $40$s using $32$ $8$-core machines. We also compared $\mathsf{Cirrus}$ with Hekaton (CCS'24), and $\mathsf{Cirrus}$ is faster when proving PLONK-friendly circuits such as Pedersen hash.

In large-scale protests, a repressive government will often disable the Internet to thwart communication between protesters. Smartphone mesh networks, which route messages over short range, possibly ephemeral, radio connections between nearby phones, allow protesters to communicate without relying on centralized Internet infrastructure. Unfortunately, prior work on mesh networks does not efficiently support cryptographically secure group messaging (a crucial requirement for protests); prior networks were also evaluated in unrealistically benign network environments which fail to accurately capture the link churn and physical spectrum contention found in realistic protest environments. In this paper, we introduce Amigo, an anonymous mesh messaging system which supports group communication through continuous key agreement, and forwards messages using a novel routing protocol designed to handle the challenges of ad-hoc routing scenarios. Our extensive simulations reveal the poor scalability of prior approaches, the benefits of Amigo's protest-specific optimizations, and the challenges that still must be solved to scale secure mesh networks to protests with thousands of participants.

Efficient realizations of succinct non-interactive arguments of knowledge (SNARKs) have gained popularity due to their practical applications in various domains. Among existing schemes, those based on error-correcting codes are of particular interest because of their good concrete efficiency, transparent setup, and plausible post-quantum security. However, many existing code-based SNARKs suffer from the disadvantage that they only work over specific finite fields.

In this work, we construct a code-based SNARK that does not rely on any specific underlying field; i.e., it is field-agnostic. Our construction follows the framework of Brakedown (CRYPTO '23) and builds a polynomial commitment scheme (and hence a SNARK) based on recently introduced expand-accumulate codes. Our work generalizes these codes to arbitrary finite fields; our main technical contribution is showing that, with high probability, these codes have constant rate and constant relative distance (crucial properties for building efficient SNARKs), solving an open problem from prior work.

As a result of our work we obtain a SNARK where, for a statement of size $M$ , the prover time is $O(M \log M )$ and the proof size is $O(\sqrt{M} )$. We demonstrate the concrete efficiency of our scheme empirically via experiments. Proving ECDSA verification on the secp256k1 curve requires only 0.23s for proof generation, 2 orders of magnitude faster than SNARKs that are not field-agnostic. Compared to the original Brakedown result (which is also field-agnostic), we obtain proofs that are 1.9–2.8$\times$ smaller due to the good concrete distance of our underlying error-correcting code, while introducing only a small overhead of 1.2$\times$ in the prover time.

During the past decade, Deep Neural Networks (DNNs) proved their value on a large variety of subjects. However despite their high value and public accessibility, the protection of the intellectual property of DNNs is still an issue and an emerging research field. Recent works have successfully extracted fully-connected DNNs using cryptanalytic methods in hard-label settings, proving that it was possible to copy a DNN with high fidelity, i.e., high similitude in the output predictions. However, the current cryptanalytic attacks cannot target complex, i.e., not fully connected, DNNs and are limited to special cases of neurons present in deep networks.

In this work, we introduce a new end-to-end attack framework designed for model extraction of embedded DNNs with high fidelity. We describe a new black-box side-channel attack which splits the DNN in several linear parts for which we can perform cryptanalytic extraction and retrieve the weights in hard-label settings. With this method, we are able to adapt cryptanalytic extraction, for the first time, to non-fully connected DNNs, while maintaining a high fidelity. We validate our contributions by targeting several architectures implemented on a microcontroller unit, including a Multi-Layer Perceptron (MLP) of 1.7 million parameters and a shortened MobileNetv1. Our framework successfully extracts all of these DNNs with high fidelity (88.4% for the MobileNetv1 and 93.2% for the MLP). Furthermore, we use the stolen model to generate adversarial examples and achieve close to white-box performance on the victim's model (95.8% and 96.7% transfer rate).

Perceptual hash functions map multimedia content that is perceptually close to outputs strings that are identical or similar. They are widely used for the identification of protected copyright and illegal content in information sharing services: a list of undesirable files is hashed with a perceptual hash function and compared, server side, to the hash of the content that is uploaded. Unlike cryptographic hash functions, the design details of perceptual hash functions are typically kept secret. Several governments envisage to extend this detection to end-to-end encrypted services by using Client Side Scanning and local matching against a hashed database. In August 2021, Apple hash published a concrete design for Client Side Scanning based on the NeuralHash perceptual hash function that uses deep learning. There has been a wide criticism of Client Side Scanning based on its disproportionate impact on human rights and risks for function creep and abuse. In addition, several authors have demonstrated that perceptual hash functions are vulnerable to cryptanalysis: it is easy to create false positives and false negatives once the design is known. This paper demonstrates that these designs are vulnerable in a weaker black-box attack model. It is demonstrated that the effective security level of NeuralHash for a realistic set of images is 32 bits rather than 96 bits, implying that finding a collision requires $2^{16}$ steps rather than $2^{48}$. As a consequence, the large scale deployment of NeuralHash would lead to a huge number of false positives, making the system unworkable. It is likely that most current perceptual hash function designs exhibit similar vulnerabilities.

This article proposes an extension for privacy-preserving applications to introduce sanctions or prohibition lists. When initiating a particular action, the user can prove, in addition to the application logic, that they are not part of the sanctions lists (one or more) without compromising sensitive data. We will show how this solution can be integrated into applications, using the example of extending Freedom Tool (a voting solution based on biometric passports). We will also consider ways to manage these lists, versioning principles, configuring the filter data set, combining different lists, and using the described method in other privacy-preserving applications.

Oxford University’s Computer Science Department (together with St John's College at Oxford) is hiring a senior Professor in Computer Science. We are looking for somebody with a world-leading research reputation with broad academic leadership.

Closing date for applications:

Contact: Head of Department, Professor Leslie Ann Goldberg (head-of-dept@cs.ox.ac.uk).

More information: https://my.corehr.com/pls/uoxrecruit/erq_jobspec_details_form.jobspec?p_id=173846

The circulant twin column parity mixer (TCPM) is a type of mixing layer for the round function of cryptographic permutations designed by Hirch et al. at CRYPTO 2023. It has a bitwise differential branch number of 12 and a bitwise linear branch number of 4, which makes it competitive in applications where differential security is required. Hirch et al. gave a concrete instantiation of a permutation using such a mixing layer, named Gaston, and showed the best 3-round differential and linear trails of Gaston have much higher weights than those of ASCON. In this paper, we first prove why the TCPM has linear branch number 4 and then show that Gaston's linear behavior is worse than ASCON for more than 3 rounds. Motivated by these facts, we aim to enhance the linear security of the TCPM. We show that adding a specific set of row cyclic shifts to the TCPM can make its differential and linear branch numbers both 12. Notably, by setting a special relationship between the row shift parameters of the modified TCPM, we obtain a special kind of mixlayer called the symmetric circulant twin column parity mixer. The symmetric TCPM has a unique design property that its differential and linear branch histograms are the same, which makes the parameter selection process and the security analysis convenient. Using the symmetric TCPM, we present two new 320-bit cryptographic permutations, namely (1) Gaston-S where we replace the mixing layer in Gaston with the symmetric TCPM and (2) SBD which uses a low-latency degree-4 S-box as the non-linear layer and the symmetric TCPM as the mixing layer. We evaluate the security of these permutations considering differential, linear and algebraic analysis, and then provide the performance comparison with Gaston in both hardware and software. Our results indicate that Gaston-S and SBD are competitive with Gaston in both security and performance.

Side-channel attacks pose a serious risk to cryptographic implementations, particularly in embedded systems. While current methods, such as test vector leakage assessment (TVLA), can identify leakage points, they do not provide insights into their root causes. We propose ARCHER, an architecture-level tool designed to perform side-channel analysis and root cause identification for software cryptographic implementations on RISC-V processors. ARCHER has two main components: (1) Side-Channel Analysis to identify leakage using TVLA and its variants, and (2) Data Flow Analysis to track intermediate values across instructions, explaining observed leaks.

Taking the binary file of the target implementation as input, ARCHER generates interactive visualizations and a detailed report highlighting execution statistics, leakage points, and their causes. It is the first architecture-level tool tailored for the RISC-V architecture to guide the implementation of cryptographic algorithms resistant to power side-channel attacks. ARCHER is algorithm-agnostic, supports pre-silicon analysis for both high-level and assembly code, and enables efficient root cause identification. We demonstrate ARCHER’s effectiveness through case studies on AES and ASCON implementations, where it accurately traces the source of side-channel leaks.

In this work, we present a new paradigm for constructing Group Authenticated Key Exchange (GAKE). This result is the first tightly secure GAKE scheme in a strong security model that allows maximum exposure attacks (MEX) where the attacker is allowed to either reveal the secret session state or the long-term secret of all communication partners. Moreover, our protocol features the strong and realistic notion of (full) perfect forward secrecy (PFS), that allows the attacker to actively modify messages before corrupting parties. We obtain our results via a series of tightly secure transformations. Our first transformation is from weakly secure KEMs to unilateral authenticated key exchange (UAKE) with weak forward secrecy (WFS). Next, we show how to turn this into an UAKE with PFS in the random oracle model. Finally, and as one of our major novel conceptual contributions, we describe how to build GAKE protocols from UAKE protocols, also in the random oracle model. We apply our transformations to obtain two practical GAKE protocols with tight security. The first is based on the DDH assumption and features low message complexity. Our second result is based on the LWE assumption. In this way, we obtain the first GAKE protocol from a post-quantum assumption that is tightly secure in a strong model of security allowing MEX attacks.

In ASIACRYPT 2019, Andreeva et al. introduced a new symmetric key primitive called the $\textit{forkcipher}$, designed for lightweight applications handling short messages. A forkcipher is a keyed function with a public tweak, featuring fixed-length input and fixed-length (expanding) output. They also proposed a specific forkcipher, ForkSkinny, based on the tweakable block cipher SKINNY, and its security was evaluated through cryptanalysis. Since then, several efficient AEAD and MAC schemes based on forkciphers have been proposed, catering not only to short messages but also to various purposes such as leakage resilience and cloud security. While forkciphers have proven to be efficient solutions for designing AEAD schemes, the area of forkcipher design remains unexplored, particularly the lack of provably secure forkcipher constructions.

In this work, we propose forkcipher design for various tweak lengths, based on a block cipher as the underlying primitive. We provide proofs of security for these constructions, assuming the underlying block cipher behaves as an ideal block cipher. First, we present a forkcipher, $\widetilde{\textsf{F}}1$, for an $n$-bit tweak and prove its optimal ($n$-bit) security. Next, we propose another construction, $\widetilde{\textsf{F}}2$, for a $2n$-bit tweak, also proving its optimal ($n$-bit) security. Finally, we introduce a construction, $\widetilde{\textsf{F}}r$, for a general $rn$-bit tweak, achieving $n$-bit security.

This article tries to offer a solution to an environmental sustainability problem using a forward-thinking approach and tries to construct a carbon footprint tracking system based on blockchain technology while also introducing tokenization intertwined with the blockchain to make everyday use as accessible and effective as possible. This effort aims to provide a solid use case for environmental sustainability and lays the groundwork of a new generation social construct where carbon footprint is a valuable unit like money next to the other important tokenized attributes a person can possibly hold. The study proposes a blockchain-based solution to store the data. Through tokenization, the transacting and sharing is facilitated. As a result, carbon footprint data can be treated as a fungible utility token. The article tries to explain how and which blockchain technology offers an effective solution to challenges in global carbon tracking systems. In this context, a use case was proposed. The critical features of the blockchain-based platform are examined. In addition, the roles of parties and user interactions within the system are detailed. In conclusion, this article proposes the adaptation of blockchain technology together with smart contracts and tokenization to the management of carbon footprints.

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.