CRYPTOSYSTEMS addresses three main challenges that revolve around improving the robustness and scalability of distributed systems using cutting-edge cryptography: (i) laying theoretical foundations for distributed algorithms using cryptography. This includes refining classical models used in the are of distributed algorithm as well as proposing entirely new ones in order to capture unaddressed issues in existing models. These challenges are addressed in WP1. (ii) designing novel algorithms for important distributed tasks such as consensus and clock synchronisation with improved scalability and robustness by leveraging cryptography. These challenges are addressed in WP2. (iii) designing novel cryptographic tools such as threshold signatures and threshold secret sharing schemes that can be used in distributed algorithms for various important tasks, in particular for generating shared randomness. These challenges are addressed in WP3.
In the following, I will summarise the progress made so far for each of these objectives.
Objective 1: New Formal Models. I have made significant progress in exploring feasibility of distributed algorithms in new fault models. Importantly, I have published two papers dealing with a generalised fault model that distinguishes two important types of benign faults from fully malicious ones. In more detail, my works consider so-called omission faults for which an attacker can selectively delete incoming or outgoing messages. These faults are simultaneously very realistic and hard to deal with, as they may keep participating with faulty behaviour in an algorithm over an extended period of time. In two works, I present algorithms with optimal fault tolerance in a model where both benign and fully malicious faults may appear. These works have appeared at TCC 2023 and TCC 2024, respectively.
Objective 2: Efficient and Robust Distributed Algorithms. Several works of mine have improved on the efficiency and robustness of distributed algorithms for variants of consensus. On one hand, I have published several works (1x FC 2024, 2x CRYPTO 2025) that significantly improve the communication, i.e. the number of bits that parties need to exchange in order to reach consensus. w in protocols for consensus. I have also shown how to improve the time complexity, i.e. the number of communication steps of consensus algorithms for the challenging setting where a majority of the parties can be corrupted. These works have appeared at EUROCRYPT 2024 and SODA 2025, respectively.
Objective 3: Cryptography for Scalable Shared Randomness. Several of my works have improved the state of the art in efficiency and robustness for important cryptographic tools related to consensus such as randomness generation (CCS 2023, CCS 2024) and jointly computing a compact signature on a message (EUROCRYPT 2024, CCS 2024, ASIACRYPT 2024, EUROCRYPT 2025, CRYPTO 2025). These works have advanced the field significantly and have produced several followup works by other researchers, as well as ongoing follow up projects that I am currently working on.
The results listed under Objective 1 required novel insights relating to self detection of omission faulty nodes. Thereafter, such a node can discontinue participating in the protocol so as to not cause further harm to parties that remain honest in the protocol.
The results listed under Objective 2 related to communication complexity involve novel techniques that are based on expander graphs as well as improved algorithms for gossiping information efficiently. Moreover, the results related to round efficiency under dishonest majority involved a translation of techniques of the honest majority setting that required novel technical insights into the problem.
For the results listed under Objective 3, several new techniques were developed to deal with an adaptive adversary that dynamically corrupts participating protocol parties based on the flow of the protocol execution. Moreover, the results related to shared randomness generation required novel definitions and proof techniques for Publicly Verifiable Secret Sharing (PVSS) that may be of independent interest.
