International Association
for Cryptologic Research

Although privately programmable pseudorandom functions (PPPRFs) are known to have numerous applications, so far, the only known constructions rely on Learning with Error (LWE) or indistinguishability obfuscation. We show how to construct a relaxed PPPRF with only one-way functions (OWF). The resulting PPPRF satisfies $1/\textsf{poly}$ security and works for polynomially sized input domains. Using the resulting PPPRF, we can get new results for preprocessing Private Information Retrieval (PIR) that improve the state of the art. Specifically, we show that relying only on OWF, we can get a 2-server preprocessing PIR with polylogarithmic bandwidth while consuming $\widetilde{O}_\lambda(N^{\frac12 + \epsilon})$ client space and $N^{1+\epsilon}$ server space for an arbitrarily small constant $\epsilon \in (0, 1)$. In the 1-server setting, we get a preprocessing PIR from OWF that achieves polylogarithmic online bandwidth and $\widetilde{O}_\lambda(N^{\frac12 + \epsilon})$ offline bandwidth, while preserving the same client and server space as before. Our result, in combination with the lower bound of Ishai, Shi, and Wichs (CRYPTO'24), establishes a tight understanding of the bandwidth and client space tradeoff for 1-server preprocessing PIR from Minicrypt assumptions. Interestingly, we are also the first to show non-trivial ways to combine client-side and server-side preprocessing to get improved results for PIR.

The Fujisaki-Okamoto transform (FO) is the go-to method for achieving chosen-ciphertext (CCA) security for post-quantum key encapsulation mechanisms (KEMs). An important step in FO is augmenting the decryption/ decapsulation algorithm with a re-encryption step -- the decrypted message is re-encrypted to check whether the correct encryption randomness was used. While solving a security problem (ciphertext-malleability), re-encryption has turned out to introduce side-channel vulnerabilities and is computationally expensive, which has lead designers to searching for alternatives. In this work, we perform a comprehensive study of such alternatives. We formalize a central security property, computational rigidity, and show that it is sufficient for obtaining CCA security. We present a framework for analyzing algorithms that can replace re-encryption and still achieve rigidity, and analyze existing proposals in this framework. Along the way, we pick up a novel QROM security statement for explicitly rejecting KEMs based on deterministic PKE schemes, something that so far only was possible when requiring a hard-to-ensure quantum property for the base PKE scheme.

The U.S. National Institute of Standards and Technology recently standardized the first set of post-quantum cryptography algo- rithms. These algorithms address the quantum threat, but also present new challenges due to their larger memory and computational footprint. Three of the four standardized algorithms are lattice based, offering good performance but posing challenges due to complex implementation and intricate security assumptions. A more conservative choice for quantum- safe authentication are hash-based signature systems. However, due to large signature sizes and low signing speeds, hash-based systems have only found use in niche applications. The first NIST standardized, state- less hash-based signature system is the SPHINCS+-based SLH-DSA. In this work we combine different approaches to show that SPHINCS+ can be optimized in its parameters and implementation, to be high per- forming, even when signing in an embedded setting. We demonstrate this in the context of user authentication using hardware security keys within FIDO. Our SPHINCS+-based implementation can even outper- form lattice-based solutions while remaining highly portable. Due to con- servative security assumptions, our solution does not require a hybrid construction and can perform authentication on current security keys. For reproducibility and to encourage further research we publish our Cortex M4-based implementation.

Threshold signatures have become a critical tool in cryptocurrency systems, offering enhanced security by distributing the signing process among multiple signers. In this work, we distribute this process between a client and a permissionless decentralized blockchain, and present novel protocols for ECDSA and EdDSA/Schnorr signatures in this setting. Typical threshold access architectures used by trusted custodians suffer from the honeypot problem, wherein the more assets the custodian holds, the greater the incentive of compromising it.

Implementing threshold signatures over permissionless blockchains poses a few challenges. First, existing networks typically work over an asynchronous reliable broadcast communication channel. Accordingly, our protocol is implemented over such a channel. As a result, it also benefits from identifiable abort, public verifiability, and guaranteed output delivery, and the client benefits from censorship resistance of blockchain systems. Second, upon signing each block, the participating quorum may dynamically change and is post-determined. Therefore, we design a fluid protocol, that supports a post-determined dynamic quorum in each communication round, thereby complying with existing broadcast channel implementations. Third, in permissionless networks, parties may join, leave, and change their stake. Therefore, we offer protocols for network reconfiguration, with complexity independent of the number of clients in the system, and our protocol efficiently supports a weighted threshold access structure for the network. Specifically, the complexity of distributed key generation and presign only depends on the number of parties and not on the overall weight, and the amortized cost of sign only depends on the individual weight.

Furthermore, our protocol introduces key improvements, including the removal of zero-knowledge proofs towards the client, and presigns with a non-interactive client. For Schnorr, the presigns are client-independent, and can be collected by the blockchain in a common pool, available for all clients in the system. These optimizations reduce communication overhead and improve the system's ability to handle traffic spikes during high-demand periods.

Our protocol is UC-secure, and is therefore natively designed for multiple clients to use the system in parallel. Notably, we propose a novel assumption, Slightly-Enhanced ECDSA Unforgeability, offering concrete security for 256-bit elliptic curves for threshold ECDSA with support for parallel execution of presigns.

In addition to securing cryptocurrency wallets, we demonstrate how our protocol enables various cross-chain applications, such as decentralized bridges, future transactions, andwallet transfer. Our system is designed for interoperability across multiple blockchains, enhancing security, scalability, and flexibility for decentralized finance (DeFi) ecosystems.

Zero-Knowledge Succinct Non-interactive Arguments of Knowledge (zkSNARKs) lead to proofs that can be succinctly verified but require huge computational resources to generate. Prior systems outsource proof generation either through public delegation, which reveals the witness to the third party, or, more preferably, private delegation that keeps the witness hidden using multiparty computation (MPC). However, current private delegation schemes struggle with scalability and efficiency due to MPC inefficiencies, poor resource utilization, and suboptimal design of zkSNARK protocols.

In this paper, we introduce DFS, a new zkSNARK that is delegation-friendly for both public and private scenarios. Prior work focused on optimizing the MPC protocols for existing zkSNARKs, while DFS uses co-design between MPC and zkSNARK so that the protocol is efficient for both distributed computing and MPC. In particular, DFS achieves linear prover time and logarithmic verification cost in the non-delegated setting. For private delegation, DFS introduces a scheme with zero communication overhead in MPC and achieves malicious security for free, which results in logarithmic overall communication; while prior work required linear communication. Our evaluation shows that DFS is as efficient as state-of-the-art zkSNARKs in public delegation; when used for private delegation, it scales better than previous work. In particular, for $2^{24}$ constraints, the total communication of DFS is less than $500$KB, while prior work incurs $300$GB, which is linear to the circuit size. Additionally, we identify and address a security flaw in prior work, EOS (USENIX'23).

Syndrome decoding (SD), and equivalently Learning Parity with Noise (LPN), is a fundamental problem in cryptography, which states that for a field $\mathbb{F}$, some compressing public matrix $\mathbf{G} \in \mathbb{F}^{k\times n}$, and a secret sparse vector $\mathbf{e} \in\mathbb{F}^{n}$ sampled from some noise distribution, $\mathbf{G}\mathbf{e}$ is indistinguishable from uniform. Recently, the SD has gained significant interest due to its use in pseudorandom correlation generators (PCGs).

In pursuit of better efficiency, we propose a new assumption called Stationary Syndrome Decoding (SSD). In SSD, we consider $q$ correlated noise vectors $\mathbf{e}_{1},\ldots,\mathbf{e}_{q}\in \mathbb{F}^n$ and associated instances $\mathbf{G}_{1}\mathbf{e}_{1},\ldots,\mathbf{G}_{q}\mathbf{e}_{q}$ where the noise vectors are restricted to having non-zeros in the same small subset of $t$ positions $L\subset [n]$. That is, for all $i\in L$, $\mathbf{e}_{j,i}$ is uniformly random, while for all other $i$, $\mathbf{e}_{j,i} = 0$.

Although naively reusing the noise vector renders SD and LPN insecure via simple Gaussian elimination, we observe known attacks do not extend to our correlated noise. We show SSD is unconditionally secure against so-called linear attacks, e.g., advanced information set decoding and representation techniques (Esser and Santini, Crypto 2024). We further adapt the state-of-the-art nonlinear attack (Briaud and Oygarden, Eurocrypt 2023) to SSD and demonstrate both theoretically and experimentally resistance to the attack.

We apply SSD to PCGs to amortize the cost of noise generation protocol. For OT and VOLE generation, each instance requires $O(t)$ communication instead of $O(t\log n)$. For suggested parameters, we observe a $1.5\times$ improvement in the running time or between 6 and $18\times$ reduction in communication. For Beaver triple generation using Ring LPN, our techniques have the potential for substantial amortization due to the high concrete overhead of the Ring LPN noise generation.

This paper introduces Neo, a new lattice-based folding scheme for CCS, an NP-complete relation that generalizes R1CS, Plonkish, and AIR. Neo's folding scheme can be viewed as adapting the folding scheme in HyperNova (CRYPTO'24), which assumes elliptic-curve based linearly homomorphic commitments, to the lattice setting. Unlike HyperNova, Neo can use “small” prime fields (e.g., over the Goldilocks prime). Additionally, Neo provides plausible post-quantum security.

Prior to Neo, folding schemes in the lattice setting, notably LatticeFold (ePrint 2024/257), worked with constraint systems defined over a cyclotomic polynomial ring. This structure allows packing a fixed batch of constraint systems over a small prime field into a single constraint system over a polynomial ring. However, it introduces significant overheads, both for committing to witnesses (e.g., the commitment scheme cannot take advantage of bit-width of values), and within the folding protocol itself (e.g., the sum-check protocol is run over cyclotomic polynomial rings). Additionally, the required ring structure places restrictions on the choice of primes (e.g., LatticeFold is not compatible with the Goldilocks field).

Neo addresses these problems, by drawing inspiration from both HyperNova and LatticeFold. A key contribution is a folding-friendly instantiation of Ajtai's commitments, with "pay-per-bit" commitment costs i.e., the commitment costs scale with the bit-width of the scalars (e.g., committing to a vector of bits is $32\times$ cheaper than committing to a vector of $32$-bit values). This scheme commits to vectors over a small prime field. It does so by transforming the provided vector into a matrix and committing to that matrix. We prove that this commitment scheme provides the desired linear homomorphism for building a folding scheme. Additionally, like HyperNova, Neo runs a single invocation of the sum-check protocol, where in HyperNova it is over the scalar field of an elliptic curve and in Neo it is over an extension of a small prime field.

Ever since the introduction of encryption, society has been divided over whether the government (or law enforcement agencies) should have the capability to decrypt private messages (with or without a war- rant) of its citizens. From a technical viewpoint, the folklore belief is that semantic security always enables some form of steganography. Thus, adding backdoors to semantically secure schemes is pointless: it only weakens the security of the “good guys”, while “bad guys” can easily circumvent censorship, even if forced to hand over their decryption keys.

In this paper we put a dent in this folklore. We formalize three worlds: Dictatoria (“dictator wins”: no convenient steganography, no user co- operation needed), Warrantland (“checks-and-balances”: no convenient steganography, but need user’s cooperation) and Privatopia (“privacy wins”: built-in, high-rate steganography, even if giving away the decryption key). We give strong evidence that all these worlds are possible, thus reopening the encryption debate on a technical level.

Our main novelty is the definition and design of special encryption schemes we call anamorphic-resistant (AR). In contrast to so called “anamorphic schemes”, — which were studied in the literature and form the basis of Privatopia, — any attempt to steganographically communicate over an AR-encryption scheme will be either impossible or hugely slow (depending on the definitional details).

Komargodski et. al. defined evolving secret-sharing schemes with an unbounded number of parties. In this model, parties arrive one after the other and the number of parties that will arrive is not known. Another cryptographic primitive related to secret-sharing is conditional disclosure of secrets protocols (CDS) that defined by Gertner et. al. A CDS protocol for a Boolean function $f$ involves $k$ servers and a referee. Each server holds a common secret $s$, a common random string $r$, and a private input $x_i$; using these $r$, each server locally computes one message and sends it to the referee. The referee, knowing the inputs $x_1, \cdots, x_k$ and the messages, should be able to compute $s$ if $f(x_1, \cdots, x_k) = 1$. Otherwise, the referee should not learn information about the secret. In a sense, this is a non-monotone version of secret sharing schemes.

Peter (ISC 23'), defines evolving CDS, implementing in particular evolving predicate $f:\{0,1\}^*\rightarrow\{0,1\}$ (he handles somewhat more general predicates for larger input domains, but generalizing to other input domains is not hard, and we focus on boolean predicates). In this setting, the number of parties is unbounded, where the parties arrive in sequential order. Each party, when arrives, sends one random message to a referee, that depending on its input, the secret, and a common randomness. He also devise evolving CDS protocols for a general evolving predicate via a black-box reduction to evolving secret-sharing scheme for a related access structure. He analyzes this construction for general access structures, as well as other classes of functions, which yields message complexity $O(2^t)$ for the worst predicates.

In this work we provide new upper and lower bounds for evolving CDS. Observing that CDS has the potential for improved efficiency, as it is not restricted to monotone operations, we devise a new evolving general CDS construction. In particular, our construction relies on representing the evolving predicate via an infinite branching program - LINBP, generalizing the monotone infinite branching program based construction of Alon et. al of evolving secret sharing schemes. We obtain nontrivial ($2^{ct-o(t)}$ for $c<1$) message complexity for branching programs of larger width than Alon et al's construction (even when restricting attention to monotone predicates), as well as Peter's construction for certain (but not all) $f$'s.

Indeed, we prove that our construction, as well as Peter's article (ISC 23’) is tight for a certain evolving predicate -- as for evolving secret-sharing, (so called strong) evolving CDS also requires share complexity of $2^{t-o(t)}$. This is unlike the state of affairs for the finite setting, where the best known CDS constructions are much more efficient than the best known secret sharing schemes (for the hardest monotone functions). The latter bound is proved based on an adaptation of Mazor's lower bound (in turns based on Csirmaz's lower bounding technique) to the CDS setting. It relies on a reduction from secret sharing for a certain class of infinite access structures -- the so called partite access structures to evolving CDS for a related (not necessarily monotone) function. Then, a partite evolving access structure is crafted using the Csirmaz-type argument.

We present the first construction for adaptively secure HIBE, that does not rely on bilinear pairings or random oracle heuristics. Notably, we design an adaptively secure HIBE from any selectively secure IBE system in the standard model. Combining this with known results, this gives the first adaptively secure HIBE system from a wide variety of standard assumptions such as CDH/Factoring/LWE/LPN. We also extend our adaptively secure HIBE system to satisfy full anonymity, giving the first adaptively secure anonymous HIBE under CDH/LWE assumption. All our HIBE systems support unbounded length identities as well as unbounded number of recursive delegation operations.

Dynamic Decentralized Functional Encryption (DDFE) is a generalization of Functional Encryption which allows multiple users to join the system dynamically without interaction and without relying on a trusted third party. Users can independently encrypt their inputs for a joint evaluation under functions embedded in functional decryption keys; and they keep control on these functions as they all have to contribute to the generation of the functional keys. In this work, we present new generic compilers which, when instantiated with existing schemes from the literature, improve over the state-of-the-art in terms of security, computational assumptions and functionality. Specifically, we obtain the first adaptively secure DDFE schemes for inner products in both the standard and the stronger function-hiding setting which guarantees privacy not only for messages but also for the evaluated functions. Furthermore, we present the first DDFE for inner products whose security can be proven under the LWE assumption in the standard model. Finally, we give the first construction of a DDFE for the attribute-weighted sums functionality with attribute-based access control (with some limitations). All prior constructions guarantee only selective security, rely on group-based assumptions on pairings, and cannot provide access control.

Over the past decade and a half, cryptanalytic techniques for Salsa20 have been increasingly refined, largely following the overarching concept of Probabilistically Neutral Bits (PNBs) by Aumasson et al. (FSE 2008). In this paper, we present a novel criterion for choosing key-$\mathcal{IV}$ pairs using certain 2-round criteria and connect that with clever tweaks of existing techniques related to Probabilistically Independent $\mathcal{IV}$ bits (earlier used for ARX ciphers, but not for Salsa20) and well-studied PNBs. Through a detailed examination of the matrix after initial rounds of Salsa20, we introduce the first-ever cryptanalysis of Salsa20 exceeding $8$ rounds. Specifically, Salsa20/$8.5$, consisting of $256$ secret key bits, can be cryptanalyzed with a time complexity of $2^{245.84}$ and data amounting to $2^{99.47}$. Further, the sharpness of our attack can be highlighted by showing that Salsa20/$8$ can be broken with time $2^{186.01}$ and data $2^{99.73}$, which is a significant improvement over the best-known result of Coutinho et al. (Journal of Cryptology, 2023, time $2^{217.14}$ and data $2^{113.14}$). Here, the refinements related to backward biases for PNBs are also instrumental in achieving the improvements. We also provide certain instances of how these ideas improve the cryptanalysis on $128$-bit versions. In the process, a few critical points are raised on some existing state-of-the-art works in this direction, and in those cases, their estimates of time and data are revisited to note the correct complexities, revising the incorrect numbers.

The wide adoption of deep neural networks (DNNs) raises the question of how can we equip them with a desired cryptographic functionality (e.g, to decrypt an encrypted input, to verify that this input is authorized, or to hide a secure watermark in the output). The problem is that cryptographic primitives are typically designed to run on digital computers that use Boolean gates to map sequences of bits to sequences of bits, whereas DNNs are a special type of analog computer that uses linear mappings and ReLUs to map vectors of real numbers to vectors of real numbers. This discrepancy between the discrete and continuous computational models raises the question of what is the best way to implement standard cryptographic primitives as DNNs, and whether DNN implementations of secure cryptosystems remain secure in the new setting, in which an attacker can ask the DNN to process a message whose "bits" are arbitrary real numbers. In this paper we lay the foundations of this new theory, defining the meaning of correctness and security for implementations of cryptographic primitives as ReLU-based DNNs. We then show that the natural implementations of block ciphers as DNNs can be broken in linear time by using such nonstandard inputs. We tested our attack in the case of full round AES-128, and had $100\%$ success rate in finding $1000$ randomly chosen keys. Finally, we develop a new method for implementing any desired cryptographic functionality as a standard ReLU-based DNN in a provably secure and correct way. Our protective technique has very low overhead (a constant number of additional layers and a linear number of additional neurons), and is completely practical.

In this article we present a non-uniform reduction from rank- 2 module-LIP over Complex Multiplication fields, to a variant of the Principal Ideal Problem, in some fitting quaternion algebra. This reduction is classical deterministic polynomial-time in the size of the inputs. The quaternion algebra in which we need to solve the variant of the principal ideal problem depends on the parameters of the module-LIP problem, but not on the problem’s instance. Our reduction requires the knowledge of some special elements of this quaternion algebras, which is why it is non-uniform. In some particular cases, these elements can be computed in polynomial time, making the reduction uniform. This is the case for the Hawk signature scheme: we show that breaking Hawk is no harder than solving a variant of the principal ideal problem in a fixed quaternion algebra (and this reduction is uniform).

We address the problem of proving the validity of computation on ciphertexts of homomorphic encryption (HE) schemes, a feature that enables outsourcing of data and computation while ensuring both data privacy and integrity. We propose a new solution that handles computations in RingLWE-based schemes, particularly the CKKS scheme for approximate arithmetic. Our approach efficiently handles ciphertext arithmetic in the polynomial ring $R_q$ without emulation overhead and manages ciphertexts maintenance operations, such as modulus switching, key switching, and rescaling, with small cost. Our main result is a succinct argument that efficiently handles arithmetic computations and range checks over the ring $R_q$. To build this argument system, we construct new polynomial interactive oracle proofs (PIOPs) and multilinear polynomial commitments supporting polynomials over $R_q$, unlike prior work which focused on finite fields. We validate the concrete complexity of our approach through implementation and experimentation. Compared to the current state-of-the-art on verifiable HE for RNS schemes, we present similar performance for small circuits while being able to efficiently scale to larger ones, which was a major challenge for previous constructions as it requires verifying procedures such as relinearization.

We investigate two natural relaxations of quantum cryptographic assumptions. First, we examine primitives such as pseudorandom generators ($\text{PRG}$s) and pseudorandom states ($\text{PRS}$s), extended with quantum input sampling, which we term $\text{PRG}^{qs}$ and $\text{PRS}^{qs}$. In these primitives, the input is sampled via a quantum algorithm rather than uniformly at random. The second relaxation, $\bot$-pseudodeterminism, allows the generator to output $\bot$ on an inverse-polynomial fraction of inputs.

We demonstrate an equivalence between (bounded-query) logarithmic-sized $\text{PRS}^{qs}$, logarithmic-sized $\text{PRS}^{qs}$, and $\text{PRG}^{qs}$. Notably, such an equivalence remains unknown for the uniform key sampling versions of these primitives. Furthermore, we establish that $\text{PRG}^{qs}$ can be constructed from $\bot$-pseudodeterministic $\text{PRG}$s ($\bot\text{-PRG}$s).

To further justify our exploration, we present two separation results. First, we examine the relationship between $\bot$-pseudodeterministic notions and their deterministic counterparts. We show that there does not exist a black-box construction of a one-way state generator $(\text{OWSG})$ from a $\bot\text{-PRG}$, indicating that $\bot$-pseudodeterministic primitives may be inherently weaker than their deterministic counterparts. Second, we explore the distinction between quantum and uniform input sampling. We prove that there does not exist a black-box construction of a $\bot$-psuedodeterministic $\text{OWSG}$ from a $\text{PRF}^{qs}$, suggesting that primitives relying on quantum input sampling may be weaker than those using traditional uniform sampling. Given the broad cryptographic applicability of $\text{PRF}^{qs}$s and $\bot\text{-PRG}$s, these separation results yield numerous new insights into the hierarchy of primitives within MicroCrypt.

Decentralized Autonomous Organization operates without a central entity, being owned and governed collectively by its members. In this organization, decisions are carried out automatically through smart contracts for routine tasks, while members vote for unforeseen issues. Scalability in decision-making through voting on proposals is essential to accommodate a growing number of members without sacrificing security. This paper addresses this challenge by introducing a scalable and secure DAO voting system that ensures security through Groth16 zk-SNARKs and exponential ElGamal encryption algorithm while achieving scalability by verifiably delegating heavy computations to untrusted entities. While offline computation on the exponential ElGamal homomorphic encryption algorithm is enabled to reduce the computational cost of the blockchain, Groth16 is allowed to maintain robust off-chain calculation without revealing any further details. Specifically, the Groth16 proof guarantees that (i) the encrypted votes accurately reflect the voter's voting power, ensuring no unauthorized weight manipulation; (ii) only valid non-negative vote values are encrypted, preventing unintended or malicious vote tampering; and (iii) the homomorphic summation is performed correctly. The implementation shows that the proofs are verified remarkably fast, making the S2DV protocol highly suitable for scalable DAO voting, while preserving the security of the election.

In this work, we consider the communication complexity of MPC protocols in honest majority setting achieving malicious security in both information-theoretic setting and computational setting. On the one hand, we study the possibility of basing honest majority MPC protocols on oblivious linear evaluation (OLE)-hybrid model efficiently with information-theoretic security. More precisely, we instantiate preprocessing phase of the recent work Sharing Transformation (Goyal, Polychroniadou, and Song, CRYPTO 2022) assuming random OLE correlations. Notably, we are able to prepare packed Beaver triples with malicious security achieving amortized communication of $O(n)$ field elements plus a number of $O(n)$ OLE correlations per packed Beaver triple, which is the best known result. To further efficiently prepare random OLE correlations, we resort to IKNP-style OT extension protocols (Ishai et al., CRYPTO 2003) in random oracle model.

On the other hand, we derive a communication lower bound for preparing OLE correlations in the information-theoretic setting based on negative results due to Damgård, Larsen, and Nielsen (CRYPTO 2019).

Combining our positive result with the work of Goyal, Polychroniadou, and Song (CRYPTO 2022), we derive an MPC protocol with amortized communication of $O(\ell+\kappa)$ elements per gate in random oracle model achieving malicious security, where $\ell$ denotes the length of a field element and $\kappa$ is the security parameter.

Several faculty positions within the Cybersecurity domain are available in Universitat Politecnica de Catalunya - BarcelonaTech (UPC), School of Engineering in Manresa. UPC is a leading research university in Spain with more than 5000 students studying on Information and Communication Technology related fields. It is the top Spanish university in terms of participation in EU-funded research projects.

Apart from conducting research within their area, the successful candidates are expected to teach the recent MSc program on Cybersecurity, AI, and IoT areas, funded through the European Commission’s Digital Europe programme. The development of the MSc program will be part of a European project, in which the candidates will participate actively.

Skills/Qualifications:

• PhD degree in the area of Computer Science, Computer Engineering, or a related field
• Written and spoken proficiency in English (Spanish or Catalan is a plus)
• Proven scientific publication record
• Experience in academic teaching (classes and lab courses) and supervising or co-supervising of academic theses

What we offer:

• A friendly working environment
• Hybrid work modality with up to 40% home office option
• Wide range of internal and external training opportunities, various career options
• Nice climate

The initial appointment will be a visiting professor position (inline with the university staff categories), during the timeframe of which the recruited professors will be given support to switch to tenure-track positions.

Closing date for applications:

Contact: Ilker Demirkol ([email protected])

Fully Homomorphic Encryption (FHE) allows computations on encrypted data without requiring decryption, ensuring data privacy during processing. However, FHE introduces a significant expansion of ciphertext sizes compared to plaintexts, which results in higher communication. A practical solution to mitigate this issue is transciphering, where only the master key is homomorphically encrypted, while the actual data is encrypted using a symmetric cipher, usually a stream cipher. The server then homomorphically evaluates the stream cipher to convert the encrypted data into a homomorphically encrypted form.

We introduce Transistor, a stream cipher specifically designed for efficient homomorphic evaluation within the TFHE scheme, a widely-used FHE framework known for its fast bootstrapping and ability to handle low-precision data. Transistor operates on $\mathbb{F}_{17}$ which is chosen to optimize TFHE performances. Its components are carefully engineered to both control noise growth and provide strong security guarantees. First, a simple TFHE-friendly implementation technique for LFSRs allows us to use such components to cheaply increase the state size. At the same time, a small Finite State Machine is the only part of the state updated non-linearly, each non-linear operation corresponding in TFHE to a rather expensive Programmable Bootstrapping. This update is done using an AES-round-like transformation. But, in contrast to other stream ciphers like SNOW or LEX, our construction comes with information-theoretic security arguments proving that an attacker cannot obtain any information about the secret key from three or fewer consecutive keystream outputs. These information-theoretic arguments are then combined with a thorough analysis of potential correlations to bound the minimal keystream length required for recovering the secret key.

Our implementation of Transistor significantly outperforms the state of the art of TFHE transciphering, achieving a throughput of over 60 bits/s on a standard CPU, all while avoiding the need for an expensive initialization process.