EN FR
EN FR
CAPSULE - 2025

2025​Activity reportProject-TeamCAPSULE​‌

RNSR: 202324388N
  • Research center​​ Inria Centre at Rennes​​​‌ University
  • In partnership with:​Université de Rennes
  • Team​‌ name: Applied Cryptography and​​ Implementation Security
  • In collaboration​​​‌ with:Institut de recherche​ en informatique et systèmes​‌ aléatoires (IRISA)

Creation of​​ the Project-Team: 2023 January​​​‌ 01

Each year, Inria​ research teams publish an​‌ Activity Report presenting their​​ work and results over​​​‌ the reporting period. These​ reports follow a common​‌ structure, with some optional​​ sections depending on the​​​‌ specific team. They typically​ begin by outlining the​‌ overall objectives and research​​ programme, including the main​​​‌ research themes, goals, and​ methodological approaches. They also​‌ describe the application domains​​ targeted by the team,​​​‌ highlighting the scientific or​ societal contexts in which​‌ their work is situated.​​

The reports then present​​​‌ the highlights of the​ year, covering major scientific​‌ achievements, software developments, or​​ teaching contributions. When relevant,​​​‌ they include sections on​ software, platforms, and open​‌ data, detailing the tools​​ developed and how they​​​‌ are shared. A substantial​ part is dedicated to​‌ new results, where scientific​​ contributions are described in​​​‌ detail, often with subsections​ specifying participants and associated​‌ keywords.

Finally, the Activity​​ Report addresses funding, contracts,​​​‌ partnerships, and collaborations at​ various levels, from industrial​‌ agreements to international cooperations.​​ It also covers dissemination​​​‌ and teaching activities, such​ as participation in scientific​‌ events, outreach, and supervision.​​ The document concludes with​​​‌ a presentation of scientific​ production, including major publications​‌ and those produced during​​ the year.

Keywords

Computer​​​‌ Science and Digital Science​

  • A1.2.8. Network security
  • A4.3.​‌ Cryptography
  • A4.3.1. Public key​​ cryptography
  • A4.3.2. Secret key​​​‌ cryptography
  • A4.3.3. Cryptographic protocols​
  • A4.6. Authentication
  • A4.8. Privacy-enhancing​‌ technologies
  • A7.1.4. Quantum algorithms​​
  • A8.5. Number theory

Other​​​‌ Research Topics and Application​ Domains

  • B6.4. Internet of​‌ things
  • B9.5.1. Computer science​​
  • B9.5.2. Mathematics
  • B9.10. Privacy​​​‌

1 Team members, visitors,​ external collaborators

Research Scientists​‌

  • Patrick Derbez [INRIA​​, HDR]
  • Aurore​​​‌ Guillevic [INRIA,​ Researcher]
  • Andre Schrottenloher​‌ [INRIA, Researcher​​]
  • Yixin Shen [​​​‌INRIA, Researcher]​

Faculty Members

  • Pierre-Alain Fouque​‌ [Team leader,​​ Univ Rennes, Professor​​​‌, HDR]
  • Daniel​ De Almeida Braga [​‌Univ Rennes, Associate​​ Professor]
  • Damien Marion​​​‌ [Univ Rennes,​ Associate Professor]

Post-Doctoral​‌ Fellows

  • Pierrick Dartois [​​INRIA, from Sep​​​‌ 2025]
  • Charles Meyer-Hilfiger​ [Univ Rennes,​‌ Post-Doctoral Fellow, from​​ Oct 2025]

PhD​​​‌ Students

  • Roderick Asselineau [​AIRBUS, CIFRE,​‌ from Jun 2025]​​
  • Mathias Boucher [Univ​​ Rennes, from Sep​​​‌ 2025]
  • Clemence Chevignard‌ [Univ Rennes]‌​‌
  • Gael Claudel [INRIA​​]
  • Mathieu Degre [​​​‌Univ Rennes]
  • Paul‌ Delhom [Univ Rennes‌​‌, CIFRE, from​​ Nov 2025]
  • Marie​​​‌ Euler [DGA-MI,‌ from Oct 2025]‌​‌
  • Baptiste Germon [Univ​​ Rennes]
  • Théo Goureau​​​‌ [INRIA, from‌ Oct 2025]
  • Aymeric‌​‌ Hiltenbrand [Univ Rennes​​]
  • Alisee Lafontaine [​​​‌INRIA, from Oct‌ 2025]
  • Phuong Nguyen‌​‌ [Univ Rennes,​​ until Feb 2025]​​​‌
  • Guilhem Niot [PQSHIELD‌, CIFRE]
  • Pierrick‌​‌ Philippe [INRIA,​​ from Oct 2025]​​​‌
  • Aurel Pichollet–Mugnier [Univ‌ Rennes]

Technical Staff‌​‌

  • Zoe Vignes [INRIA​​, Engineer, from​​​‌ Sep 2025]

Interns‌ and Apprentices

  • Alexandre Autran‌​‌ [INRIA, Intern​​, from Jun 2025​​​‌ until Aug 2025]‌
  • Mathias BOUCHER [Univ‌​‌ Rennes, from Mar​​ 2025 until Jun 2025​​​‌]
  • Todd Cauet-Male [‌INRIA, Intern,‌​‌ until Jul 2025]​​
  • Hubert DE GROOTE [​​​‌Univ Rennes, from‌ Mar 2025 until Aug‌​‌ 2025]
  • Alisée LAFONTAINE​​ [INRIA, from​​​‌ Mar 2025 until Aug‌ 2025]
  • Adrien LAGASSE‌​‌ [INRIA, from​​ May 2025 until Aug​​​‌ 2025]
  • Babacar Ndiaye‌ [INRIA, Intern‌​‌, from Apr 2025​​ until Aug 2025]​​​‌

Administrative Assistants

  • Loïc LESAGE‌ [INRIA, from‌​‌ Mar 2025 until Sep​​ 2025]
  • Eleonora SABA​​​‌ [INRIA, from‌ Oct 2025]

External‌​‌ Collaborators

  • Clément Dell'Aiera [​​DGA-MI]
  • Marie Euler​​​‌ [DGA-MI]
  • Mathieu‌ Goessens [Univ Rennes‌​‌, until Aug 2025​​]
  • Tuong-Huy Nguyen [​​​‌DGA-MI]

2 Overall‌ objectives

Nowadays, and contrary‌​‌ to the past decades,​​ the design of cryptographic​​​‌ algorithms follows an integrated‌ approach which considers security,‌​‌ efficiency and implementation requirements​​ at the same time.​​​‌ The research activities of‌ the team CAPSULE tackle‌​‌ these challenges in order​​ to provide more secure​​​‌ cryptographic implementations and applications‌ deployed in the real‌​‌ world.

  • Highly efficient symmetric​​ cryptosystems are a prerequisite​​​‌ for all cryptographic infrastructure.‌ Recently, many new designs‌​‌ have been proposed, which​​ aim to perform well​​​‌ under various constraints (e.g.,‌ lightweight cryptographic schemes, or‌​‌ schemes tailored for advanced​​ FHE and MPC protocols).​​​‌ The confidence in these‌ schemes is based on‌​‌ cryptanalysis, analyzing their security​​ against classical and quantum​​​‌ adversaries. Our research lies‌ not only in finding‌​‌ new attacks, but also​​ in designing automated audit​​​‌ tools that simplify and‌ systematize this task.
  • Post-quantum‌​‌ security is a major​​ challenge that cryptographers are​​​‌ facing right now. As‌ new post-quantum designs for‌​‌ encryption and digital signatures​​ are being standardized by​​​‌ NIST, the CAPSULE team‌ is actively involved in‌​‌ further improving the efficiency​​ of these schemes and​​​‌ their security analysis, both‌ against classical and quantum‌​‌ adversaries.
  • Both symmetric and​​ asymmetric cryptosystems need ultimately​​​‌ to be implemented, and‌ these implementations can be‌​‌ vulnerable to various types​​ of side-channel attacks. Finding​​​‌ new attacks and implementing‌ new countermeasures are two‌​‌ sides of the same​​​‌ coin.
  • We are also​ interested in studying the​‌ security of well-known deployed​​ systems, such as the​​​‌ security of TLS, secure​ messaging, and databases.

3​‌ Research program

3.1 Security​​ against post-quantum attackers

The​​​‌ seminal paper of Peter​ Shor at FOCS 1994​‌ 97 shows that if​​ we were able to​​​‌ build quantum computers, then​ the factorization and discrete​‌ logarithm problems could be​​ solved in polynomial time.​​​‌ Since then, there has​ been a tremendous effort​‌ in the cryptographic community​​ to propose cryptosystems that​​​‌ are secured in the​ presence of quantum computers.​‌ Many alternatives to the​​ two number theoretic problems​​​‌ above have been proposed.​ Among them, our team​‌ already has activities and​​ interests in two types​​​‌ of assumptions:

  • lattice-based schemes,​ where security is based​‌ on the difficulty of​​ computing short vectors in​​​‌ random euclidean lattices;
  • code-based​ schemes, where security is​‌ based on the difficulty​​ on computing low hamming​​​‌ weight words in random​ codes.

Euclidean lattices are​‌ discrete subgroups of ℝ​​n, while codes​​​‌ are linear subspaces of​ a vector space over​‌ a finite field. The​​ semantic similarities between the​​​‌ hardness assumptions are not​ unexpected: lattices and codes​‌ appearing in cryptography are​​ often related objects, that​​​‌ one could say considered​ from different metric perspectives.​‌

In post-quantum cryptography, lattice-based​​ assumptions take an important​​​‌ place and received an​ increasing amount of attention​‌ in the last decade,​​ thanks to the strong​​​‌ security guarantees provided by​ these assumptions as well​‌ as their flexibility for​​ cryptographic designs. Indeed, Ajtai​​​‌ and Regev presented reductions​ between, respectively, finding Short​‌ Integer Solutions of random​​ linear systems (SIS) or​​​‌ solving random noisy linear​ systems (“Learning With Errors”,​‌ LWE) and computing short​​ vectors in euclidean lattices​​​‌ in the worst case.​ They both serve as​‌ the foundation of security​​ to design public-key encryptions,​​​‌ digital signatures, zero-knowledge proof​ systems, key-encapsulation mechanisms, homomorphic​‌ encryption ... In order​​ to improve practical efficiency,​​​‌ "structured" versions of these​ problems relying on lattices​‌ with symmetries have been​​ proposed. Such lattices are​​​‌ related to algebraic objects​ appearing in the geometry​‌ of numbers and some​​ of the resulting schemes​​​‌ have been the clear​ winners of NIST's call​‌ for standardization.

Better Reductions.​​

Our trust in the​​​‌ hardness of lattice-based constructions​ relies fundamentally on our​‌ understanding of the security​​ reductions between the (many,​​​‌ structured) variants of SIS​ and LWE. Depending on​‌ the additional structure allowed​​ to the designer, they​​​‌ are associated to number​ rings, ideals, and, more​‌ generally, modules over the​​ integer ring of a​​​‌ number field, and related​ to the corresponding class​‌ of lattices with symmetries.​​ Additionally, for LWE the​​​‌ noise distribution is also​ a parameter of the​‌ problem. Overall, this leads​​ to a plethora of​​​‌ variants and versions that​ need some hierarchizing and​‌ a better understanding of​​ the interplay between their​​​‌ related parameters. Thankfully, important​ classifying works have already​‌ been presented, regularly involving​​ members of our team​​​‌ (e.g. 57, 94​, 52.

Yet,​‌ there are still many​​ unclear results or relations​​ that are not yet​​​‌ satisfyingly understood. For example,‌ the fundamental reductions of‌​‌ Ajtai and Regev are​​ far from tight, incurring​​​‌ a blowup in important‌ parameters (sometimes estimated to‌​‌ be in O(​​n11)).​​​‌ While this is not‌ a problem asymptotically, it‌​‌ clearly raises concerns on​​ how to select parameters​​​‌ and the level of‌ security they actually achieve.‌​‌ However, these proofs techniques​​ have not been updated​​​‌ since their presentations: it‌ is not unlikely that‌​‌ more recent tools could​​ lead to improvements. In​​​‌ another example, there seems‌ to be a non‌​‌ smooth gap of difficulty​​ between the hardness of​​​‌ very structured variants of‌ LWE (linked to “ideal‌​‌ lattices problems”) and less-but-still-quite​​ structured ones. Roughly speaking,​​​‌ the former seems to‌ belong to subexponential complexity‌​‌ while the latter variants​​ are still considered exponential.​​​‌ Our current knowledge is‌ also not enough to‌​‌ guarantee the actual existence​​ of this gap, which​​​‌ prevents an accurate understanding‌ of the underlying problems'‌​‌ concrete hardness. In a​​ last example, one can​​​‌ also notice that all‌ the proof strategies for‌​‌ these general reductions rely​​ on the same high-level​​​‌ arguments. Yet, multiple works‌ dealing with subcases had‌​‌ to be presented to​​ reach the current state​​​‌ of the art. On‌ the one hand, it‌​‌ could be that there​​ is a unifying, all-encompassing​​​‌ presentation that would greatly‌ simplify the state of‌​‌ the affairs and bring​​ a kind of maturity​​​‌ to this field. On‌ the other hand, there‌​‌ may be fundamental obstructions​​ to a general framework,​​​‌ and highlighting them would‌ definitely help the community's‌​‌ understanding. These three examples​​ raise important questions first​​​‌ about security, but also‌ about our way of‌​‌ using the mathematical tools​​ behind these results. Our​​​‌ team's objectives are to‌ investigate all these paths‌​‌ and to find either​​ positive or negative answers​​​‌ to improve the general‌ understanding of the area.‌​‌

Algorithms for hard problems​​ and attacks on cryptosystems.​​​‌

We have proposed some‌ algorithms to study the‌​‌ security of hard computational​​ problems in cyclotomic fields​​​‌ as the Principal Ideal‌ Problem (PIP) in 48‌​‌, reducing module lattices​​ as a generalization of​​​‌ the LLL algorithm in‌ the ring of integers‌​‌ of a number field​​ in 83 or in​​​‌ a tower of cyclotomic‌ fields in 78.‌​‌ We generalized the BKW​​ algorithm to binary LWE​​​‌ setting in 79 and‌ studied the Learning Parities‌​‌ with Noise (LPN) Problem​​ in 84.

We​​​‌ have also attacked concrete‌ cryptographic schemes. We broke‌​‌ some multivariate schemes such​​ as the SFLASH signature​​​‌ schemes in 65 and‌ variants 71, and‌​‌ the ASASA schemes in​​ 87. We have​​​‌ also broken FHE schemes‌ based on overstretched NTRU‌​‌ parameters in 80 or​​ concrete FHE in 59​​​‌.

We want to‌ study the resistance of‌​‌ post-quantum cryptosystems and hard​​ problems against classical and​​​‌ quantum adversaries. It is‌ particularly interesting for lattice‌​‌ problems since the cryptanalysis​​ of these problems is​​​‌ very young. One key‌ objective in this line‌​‌ of research would be​​​‌ to find an analog​ of the BKZ algorithm​‌ for structured lattices defined​​ over a number field.​​​‌ It is also interesting​ to improve the recent​‌ work of 46,​​ which suggests that this​​​‌ problem may be weaker​ than previously thought.

Constructions​‌ and practical cryptosystems.

Applications​​ of cryptography usually culminate​​​‌ with the description of​ an efficient cryptosystem. An​‌ important part of our​​ activity in post-quantum cryptography​​​‌ therefore targets the design​ of new schemes resistant​‌ to quantum attackers, providing​​ advanced functionalities to its​​​‌ users, without sacrificing efficiency.​

In this area, members​‌ of CAPSULE have worked​​ on the lattice-based signature​​​‌ scheme Falcon and its​ efficiency-security trade-off ModFalcon 60​‌. A first objective​​ would be to extend​​​‌ in a useful way​ the so-called “trapdoor generation”​‌ which is core to​​ the two schemes above.​​​‌ In a nutshell, the​ secret key corresponds to​‌ a basis of short​​ vectors of a lattice,​​​‌ that only the user​ should be able to​‌ compute efficiently. ModFalcon already​​ extended the class of​​​‌ lattices for which this​ can be done, and​‌ it is an interesting​​ question to manage an​​​‌ even larger class of​ lattice. In terms of​‌ applications, this would allow​​ for even more flexibility,​​​‌ which can be particularly​ useful when the signature​‌ scheme is used as​​ a black box inside​​​‌ a larger cryptographic algorithm.​ It could also allow​‌ for other functionalities such​​ as threshold signatures or​​​‌ maybe masked signatures. On​ this line of thought,​‌ we are also interested​​ in designing masked lattice​​​‌ signatures or even multi-party​ signatures. While there have​‌ been very recent proposals​​ (relying on a different​​​‌ paradigm than the Falcon​ family), the efficiency is​‌ still lacking in practice.​​ A success here could​​​‌ lead to concrete industrial​ applications.

But this is​‌ not the only construction​​ on which the team​​​‌ is currently working. There​ are many interesting cryptographic​‌ constructions that need to​​ be studied to obtain​​​‌ efficient post-quantum schemes, such​ as signatures and zero-knowledge​‌ proofs, but also signatures​​ with more properties like​​​‌ group signatures, blind signatures​ ... and applications like​‌ e-voting. Indeed, a lot​​ of progress has been​​​‌ made to obtain efficient​ signatures and public key​‌ encryptions, especially with the​​ NIST competition, but the​​​‌ efficiency of more advanced​ schemes is still far​‌ from existing (but not​​ post-quantum) solutions. One of​​​‌ the big challenges would​ be to obtain efficient​‌ zero-knowledge proof systems, as​​ this primitive is often​​​‌ an easy way to​ build more advanced primitives.​‌

3.2 Symmetric Cryptography

Despite​​ being one of the​​​‌ oldest forms of cryptography,​ symmetric cryptography is a​‌ very active research area,​​ with recent activity focusing​​​‌ on new designs optimized​ for specific operational constraints.​‌ For example, the lightweight​​ cryptography competition launched by​​​‌ the NIST1 in​ 2017 concluded in 2023​‌ by selecting the lightweight​​ cipher family Ascon 64​​​‌, optimized for hardware​ implementations. At the same​‌ time, many new ciphers​​ have been proposed which​​​‌ are optimized to be​ integrated in advanced cryptographic​‌ protocols, such as the​​ FHE-friendly block cipher LowMC,​​ or protected hardware implementations.​​​‌

The team CAPSULE studies‌ the security of symmetric‌​‌ primitives such as block​​ ciphers, stream ciphers and​​​‌ hash functions, against various‌ types of attacks. We‌​‌ consider both classical and​​ quantum security, the latter​​​‌ being a prerequisite for‌ post-quantum cryptography architectures.

Tools‌​‌ for discovering new attacks.​​

Symmetric cryptosystems are widely​​​‌ used because they are‌ the only ones that‌​‌ can achieve some major​​ functionalities such as high-speed​​​‌ or low-cost encryption, fast‌ message authentication, and efficient‌​‌ hashing. But, unlike public-key​​ cryptographic algorithms, secret-key primitives​​​‌ do not have satisfying‌ security proofs. The security‌​‌ of these algorithms is​​ empirically established by cryptanalysis.​​​‌

It is obvious that‌ this security criterion, despite‌​‌ its success so far,​​ is not completely satisfactory.​​​‌ For instance we may‌ estimate that, for a‌​‌ given primitive, no more​​ than a few dozens​​​‌ of researchers are actively‌ working on breaking it.‌​‌ Hence, due to this​​ weak effort, the non-discovery​​​‌ of an attack against‌ a particular primitive does‌​‌ not mean so much.​​ Besides, finding the best​​​‌ attacks on a given‌ design is a time-consuming‌​‌ work, and errors can​​ lead to under- or​​​‌ over-estimating its security.

Therefore,‌ our team specializes in‌​‌ building tools for automatically​​ finding large classes of​​​‌ attacks. This transforms the‌ statement “we did not‌​‌ find any attack of​​ this kind”, which is​​​‌ only a subjective guarantee,‌ into “the audit tool‌​‌ X did not find​​ any attack”, which is​​​‌ a formal statement, giving‌ a quantifiable objective guarantee.‌​‌

In the past, the​​ members of the team​​​‌ have proposed many tools,‌ for example for improving‌​‌ attacks on round-reduced versions​​ of AES 54,​​​‌ Demirci-Selçuk attacks on AES‌ 63, and impossible‌​‌ differential attacks 62.​​

Our more recent work​​​‌ uses tools based on‌ MILP (Mixed Integer Linear‌​‌ Programming), SAT (Satisfiability) or​​ CP (Constraint Programming). In​​​‌ this setting, the search‌ and optimization of an‌​‌ attack are reduced to​​ a problem of a​​​‌ specific form, for which‌ an off-the-shelf solver is‌​‌ used. Besides the actual​​ work of implementing this​​​‌ reduction, our research aims‌ at better understanding the‌​‌ differences between these optimization​​ tools, finding which ones​​​‌ are more adapted for‌ a given problem, and‌​‌ adapting some of these​​ general-purpose software tools to​​​‌ particular cryptographic problems.

Finding‌ and optimizing a cryptanalytic‌​‌ attack in its entirety​​ is an especially interesting​​​‌ problem, since it requires‌ the integration of different‌​‌ steps (for example a​​ good distinguisher and a​​​‌ key-recovery phase). Since the‌ search space is of‌​‌ exponential size, often making​​ the problem intractable, it​​​‌ is possible to first‌ find an approximation of‌​‌ the best attacks and​​ then instantiate precisely the​​​‌ values of the parameters.‌ Also, if MILP, SAT‌​‌ and CP tools quickly​​ give an answer, it​​​‌ is tempting to build‌ ad-hoc tools that can‌​‌ more efficiently take into​​ account the weaknesses discovered​​​‌ by these tools.

Finally,‌ there are only a‌​‌ few tools for analyzing​​ the security of ARX​​​‌ ciphers based on additions,‌ rotations and xor operations.‌​‌ These functions are hard​​​‌ to analyze with the​ current cryptanalytic techniques, and​‌ no attack has really​​ endangered the full Chacha​​​‌ stream cipher proposed by​ Dan Bernstein or the​‌ block cipher Speck proposed​​ by the NSA. They​​​‌ can be implemented very​ efficiently in x86 processors​‌ and currently Chacha is​​ in the most used​​​‌ ciphersuites on TLS, making​ them prominent targets for​‌ cryptanalysis.

New Designs.

Our​​ goal is to analyze​​​‌ the security of the​ new symmetric-key designs by​‌ developing new cryptanalytic techniques.​​ The LowMC block cipher​​​‌ is one of the​ first symmetric primitives designed​‌ for taking into account​​ the efficiency constraints of​​​‌ public-key cryptosystems. It has​ been built as a​‌ FHE-friendly cipher, by minimizing​​ the number of multiplicative​​​‌ gates which are the​ main efficiency bottleneck for​‌ this application. Several attacks​​ have been proposed on​​​‌ LowMC and LowMC v2.​ LowMC v3 was used​‌ in Picnic, a Zero-Knowledge-based​​ post-quantum signature scheme proposed​​​‌ at the NIST competition,​ which wasn't standardized.

The​‌ Keccak hash function has​​ been standardized in 2015​​​‌ as SHA-3. Keccak brought​ new interest in a​‌ new design called Sponge​​ function and permutation-based primitives.​​​‌ Some round-reduced versions of​ SHA-3 have been used​‌ in many constructions from​​ Pseudo-Random Generator in SHAKE,​​​‌ to the Pseudo-Random Function​ Farfalle 47, the​‌ authenticated encryption scheme Keyak,​​ or the hash function​​​‌ KangarooTwelve proposed as an​ RFC. Only a few​‌ attacks have been proposed​​ against SHA-3 and new​​​‌ cryptanalysis tools need to​ be designed.

Quantum Cryptanalysis.​‌

Since 2016, many works​​ have been done in​​​‌ the cryptanalysis of symmetric​ primitives using quantum algorithms.​‌ While symmetric cryptosystems are​​ generally believed to hold​​​‌ well against adversaries equipped​ with a quantum computer,​‌ these works have substantiated​​ these claims with dedicated​​​‌ security analyses, such as​ the best attacks against​‌ reduced-round versions of the​​ standard AES 50.​​​‌

Grover's search algorithm, which​ can provide a quadratic​‌ speedup on exhaustive key​​ search (from 2k​​​‌ operations to 2k​/2), is​‌ often cited as the​​ main player in the​​​‌ quantum security of symmetric​ primitives. However, in the​‌ past few years, the​​ landscape of quantum algorithms​​​‌ for cryptanalysis has considerably​ expanded, with notable results​‌ such as quantum speedups​​ above quadratic for specific​​​‌ constructions 51. These​ recent works highlight the​‌ benefit of combining state-of-the-art​​ quantum algorithms and symmetric​​​‌ cryptanalysis techniques.

In team​ CAPSULE, our research in​‌ quantum cryptanalysis is three-fold.​​

First, we develop new​​​‌ quantum algorithms for cryptanalytic​ problems, which we aim​‌ to apply in symmetric​​ cryptography, but may also​​​‌ have applications in public-key​ cryptography. An example of​‌ such a double-edged sword​​ is our recent work​​​‌ on quantum walks 49​.

Second, we analyze​‌ existing classical cryptanalysis techniques​​ and study how to​​​‌ translate them into quantum​ cryptanalysis techniques. Intuitively, a​‌ primitive that is classically​​ vulnerable should be quantumly​​​‌ broken as well, but​ this is not always​‌ the case, as classical​​ attack strategies are not​​​‌ always exploitable in the​ quantum setting. Our research​‌ in this area focuses​​ on the strategies which​​ can exhibit the largest​​​‌ quantum speedups, quadratic (like‌ Grover's search) or even‌​‌ above by using advanced​​ frameworks.

Finally, after identifying​​​‌ new classes of quantum‌ attacks, we aim at‌​‌ integrating these attacks into​​ automated tools. Indeed, the​​​‌ task of finding and‌ optimizing quantum attacks can‌​‌ be even more challenging​​ that classical ones, since​​​‌ they rely often on‌ different strategies, sometimes counterintuitive.‌​‌ Furthermore, since the resulting​​ procedures are quantum algorithms,​​​‌ the analysis of their‌ time and memory complexities‌​‌ comes with specific technicalities.​​ Our goal is to​​​‌ automatize this step as‌ well in a way‌​‌ that may benefit cryptanalysts​​ interested in this topic​​​‌ but unfamiliar with quantum‌ algorithms.

3.3 Elliptic curves‌​‌ for public-key cryptography

With​​ Aurore Guillevic joining the​​​‌ team in 2024, the‌ research themes extended to‌​‌ elliptic curve cryptography. In​​ public-key cryptography, elliptic curves​​​‌ over finite fields are‌ a mathematical algebraic structure‌​‌ which provides the best​​ trade-off between speed and​​​‌ key-sizes. The group of‌ points on the curve‌​‌ efficiently replaces the multiplicative​​ subgroup of prime finite​​​‌ fields as an implementation‌ choice for discrete-logarithm based‌​‌ protocols. More recently with​​ the rise of proof​​​‌ systems, elliptic curves with‌ dedicated properties are designed.‌​‌ In particular, pairing-friendly elliptic​​ curves are equipped with​​​‌ a bilinear pairing (like‌ a scalar product) that‌​‌ allows to multiply once​​ secret scalars “in the​​​‌ exponents” without revealing them.‌ It led to Succinct‌​‌ Non-interactive ARguments of Knowledge​​ (SNARK), a mechanism that​​​‌ blindly checks the validity‌ of a quadratic equation‌​‌ “in the exponents”. The​​ cornerstone work by Groth​​​‌ in 2016 obtained a‌ SNARK of the smallest‌​‌ cost in terms of​​ pairing computation and allowed​​​‌ the development of many‌ variants tailored for various‌​‌ proof systems. The work​​ in the team includes​​​‌ designing new dedicated and‌ secure elliptic curves (finding‌​‌ parameters of cryptographic size),​​ studying the security of​​​‌ existing curves, and developing‌ software modules implementing fast‌​‌ pairings on new elliptic​​ curves.

3.4 Security of​​​‌ cryptographic implementation and Real-World‌ Cryptography

In this research‌​‌ axis, our aim is​​ to study the security​​​‌ of implementations against various‌ side channels such as‌​‌ fault attacks, power analysis​​ and electromagnetic emanations, as​​​‌ well as timing attacks‌ on various cryptographic schemes‌​‌ deployed in real-world systems.​​ We are also interested​​​‌ in providing security proofs‌ for real-world systems or‌​‌ improving their security.

Hardware​​ and embedded implementations.

Side​​​‌ Channel Attacks (SCA) rely‌ on statistical tools to‌​‌ extract the secret information​​ from leakage traces. Then,​​​‌ algorithmic techniques usually based‌ on previous cryptanalytic results‌​‌ are used to efficiently​​ recover secret data. Indeed,​​​‌ the known black-box attacks‌ are extended by exploiting‌​‌ the leakage information, that​​ gives more information on​​​‌ the internal secret variables,‌ a.k.a. the grey-box model.‌​‌ The SCA information can​​ be for instance the​​​‌ Hamming weight of a‌ limited number of variables.‌​‌ Recently, the white-box model​​ has been proposed, where​​​‌ the adversary can stop‌ the execution of a‌​‌ process and has access​​ to all variables.

Side-channel​​​‌ attacks have been successfully‌ applied to break many‌​‌ embedded implementations these last​​​‌ 20 years. After the​ information theoretic approach of​‌ Ishai, Sahai and Wagner​​ 76 to prove the​​​‌ security of implementations, secure​ theoretical foundations have been​‌ laid by Prouff and​​ Rivain and later Duc​​​‌ et al. in 93​, 66. Soon​‌ after, some tools have​​ been developed such as​​​‌ 42, 43,​ 41 to protect software​‌ and hardware implementations with​​ masking techniques. Nowadays, we​​​‌ have sound masking schemes.​ Some of them already​‌ have been introduced into​​ lattice-based implementations 44,​​​‌ where generally securing randomness​ presents an interesting challenge.​‌ We aim at extending​​ the results of 44​​​‌, 45, 86​, 70 to other​‌ post-quantum alternatives like code-based,​​ multivariate, or hash-based schemes​​​‌ and to provide secure​ implementations.

More recently,​‌ other tools coming from​​ statistical learning (such as​​​‌ deep learning) have been​ proposed to break embedded​‌ implementations. They open the​​ door to powerful techniques​​​‌ and more efficient attacks.​ Template attacks model the​‌ leakage distribution with a​​ Gaussian distribution, approximating the​​​‌ actual distribution by considering​ its mean and its​‌ standard deviation. More standard​​ attacks, a.k.a. Differential Power​​​‌ Analysis (DPA), only consider​ the mean. However, higher​‌ moments can be useful​​ to consider. Deep learning​​​‌ techniques are useful to​ efficiently extract complex relations​‌ between variables even in​​ the presence of noise.​​​‌ Taking into account these​ more powerful deep learning​‌ or white-box attacks as​​ well as developing countermeasures​​​‌ is a hot, trendy​ topic in SCA. In​‌ the former, deep learning​​ allow to find correlations​​​‌ between many points of​ interest of one curve,​‌ a.k.a. horizontal attacks. In​​ the latter, white-box cryptography​​​‌ provides the adversary with​ the same kind of​‌ information, since they can​​ stop the execution of​​​‌ the program and get​ noiseless information on all​‌ of its variables. Taking​​ into account such powerful​​​‌ attackers is one main​ challenge for side-channel attacks.​‌

Finally, we are interested​​ in working on the​​​‌ new micro-architectural attacks HertzBleed​ and others. These​‌ attacks show that side-channel​​ attacks are also a​​​‌ threat to software implementations.​ Porting to software some​‌ of the many techniques​​ used to secure embedded​​​‌ systems is thus a​ major topic.

Software implementations.​‌

Constant-time implementation is a​​ programming principle that aims​​​‌ at providing code where​ the running time and​‌ memory accesses are independent​​ of the secret values.​​​‌ Timing leakage can be​ used to mount attacks​‌ on computers and smartphones.​​ There exist many tools​​​‌ in the literature that​ help developers to avoid​‌ these leakage, but insecure​​ implementations are still aplenty.​​​‌ For instance, we recently​ broke the WPA-3 implementation​‌ used in FreeRadius and​​ iwd (iNet Wireless Daemon)​​​‌ 55, and also​ found other weaknesses.

We​‌ want to discover new​​ attacks in open-source libraries​​​‌ and to help developers​ in order to verify​‌ the constant-time property of​​ their codes. For example,​​​‌ some tools are tailored​ to small pieces of​‌ cryptographic codes and do​​ not scale well with​​​‌ more complex codes that​ rely on many libraries.​‌ Our goal is to​​ provide verification tools for​​ analyzing the constant-time property​​​‌ of large source codes.‌ We are also interested‌​‌ in studying the security​​ of DRM systems used​​​‌ in widely deployed systems.‌ We do not have‌​‌ permanent researchers on reverse-engineering,​​ but we work with​​​‌ postdoc students such as‌ Alexandre Gonzalvez, as well‌​‌ as Mohamed Sabt from​​ the Spicy team on​​​‌ this topic. Besides, we‌ co-supervise 3 theses on‌​‌ the security of software​​ implementations.

Security Proofs of​​​‌ Protocols and Real-World Systems.‌

We are interested in‌​‌ studying the security of​​ cryptographic protocols deployed in​​​‌ the real-world such as‌ WhatsApp, middlebox, Content-Delivery Network‌​‌ (CDN), TLS, and 5G​​ networks. Recently, we have​​​‌ also considered the security‌ of searchable symmetric encryption,‌​‌ where the goal is​​ to outsource the storage​​​‌ of a database to‌ an untrusted server, while‌​‌ maintaining search capabilities. This​​ last area is a​​​‌ nice application of secure‌ computations and the PhD‌​‌ thesis of R. Bost​​ (P.A. Fouque's PhD student)​​​‌ in this domain received‌ the GDR Security price‌​‌ of the best PhD​​ in 2018. We also​​​‌ work with Cristina Onete,‌ an assistant professor at‌​‌ Limoges on this topic.​​ Currently, we are interested​​​‌ to propose hybridization techniques‌ between pre- and post-quantum‌​‌ cryptography for various protocols​​ such as Signal, IPSEC,​​​‌ ... in the PEPR‌ post-quantum cryptography.

Microarchitectural attacks.‌​‌

Microarchitectural attacks are very​​ potent side-channel attacks that​​​‌ exploit the microarchitecture features‌ of modern processors, such‌​‌ as caches, branch predictors,​​ and speculative execution. They​​​‌ represent a significant threat‌ to the security of‌​‌ cryptographic implementations, as they​​ can leak sensitive information​​​‌ through subtle timing variations‌ and other side effects.‌​‌ Most work focuses on​​ x86 architectures, but ARM​​​‌ architectures are also concerned,‌ especially with the widespread‌​‌ use of ARM in​​ mobile devices, and recent​​​‌ interest in deploying more‌ ARM CPUs in laptop‌​‌ and desktop computer. We​​ are interested in studying​​​‌ complex ARM microarchitectures, such‌ as the Qualcomm Snapdragon‌​‌ chip, to understand their​​ vulnerabilities to microarchitectural attacks.​​​‌ This line of work‌ involves reverse-engineering the microarchitecture,‌​‌ developing new attack techniques,​​ and proposing effective countermeasures​​​‌ to mitigate these threats.‌ We investigate these topics‌​‌ in collaboration with the​​ SUSHI Inria team and​​​‌ the ANSSI.

4 Application‌ domains

4.1 Designing, Analyzing‌​‌ and Choosing Cryptographic Standards​​

The research community is​​​‌ strongly involved in the‌ development and evolution of‌​‌ cryptographic standards. Many standards​​ are developed through open​​​‌ competitions (e.g. AES, SHA-3)‌ where multiple teams propose‌​‌ new designs, and a​​ joint cryptanalysis effort allows​​​‌ to select the most‌ suitable proposals. The analysis‌​‌ of established standards is​​ also an important work,​​​‌ in order to deprecate‌ weak algorithms before they‌​‌ can be exploited. Several​​ members of the team​​​‌ have been involved in‌ this type of effort‌​‌ and we plan to​​ continue this work to​​​‌ ensure that secure algorithms‌ are widely available. We‌​‌ believe that good cryptographic​​ standards have a large​​​‌ socio-economic impact; thus, we‌ are active in proposing‌​‌ schemes to future competitions,​​ and in analyzing schemes​​​‌ proposed to current or‌ future competitions, as well‌​‌ as widely-used algorithms and​​​‌ standards. At the moment,​ we are involved in​‌ the two standardization efforts​​ run by NIST for​​​‌ post-quantum cryptography and lightweight​ cryptography, and other real-world​‌ protocols.

NIST post-quantum competition.​​

The NIST post-quantum competition​​​‌ aims at standardizing quantum-safe​ public-key primitives. The goal​‌ is to propose a​​ quantum-safe alternative for the​​​‌ schemes based on number​ theory which are threatened​‌ by the advent of​​ quantum computers. It is​​​‌ expected to have a​ huge and long-term impact​‌ on all public-key cryptography.​​ It received 69 proposals​​​‌ in November 2017. The​ Falcon signature scheme, co-designed​‌ by some members of​​ the Capsule team, has​​​‌ been selected by NIST​ in July 2022. We​‌ have also submitted Solmae​​ to the Korean Post-Quantum​​​‌ Competition, which is a​ variant of Falcon that​‌ is easier to implement​​ hence to protect from​​​‌ SCA. Finally, we have​ also proposed BAT 69​‌, an encryption scheme​​ that follows the design​​​‌ rationale of Falcon. We​ plan to submit this​‌ scheme to the IETF​​ as it enjoys interesting​​​‌ properties in terms of​ bandwidth, that are not​‌ displayed by NIST's selected​​ key encapsulation scheme, Kyber.​​​‌

In June 2023, we​ have submitted the PROV​‌ and VOX signature schemes​​ to NIST's new call​​​‌ for digital signatures. These​ two schemes are based​‌ on multivariate cryptography problems,​​ and are variants of​​​‌ the unbalanced Oil-and-Vinegar signature​ schemes, proposed in 1997​‌ by Patarin. PROV has​​ a security proof, while​​​‌ VOX is a stronger​ version of UOV that​‌ avoids known weaknesses (namely,​​ UOV has a large​​​‌ set of isotropic vectors​ common to all quadratic​‌ forms of the public​​ key).

NIST competition on​​​‌ lightweight symmetric encryption.

The​ NIST lightweight cryptography standardization​‌ process is an initiative​​ to develop and standardize​​​‌ new authenticated encryption algorithms​ suitable for constrained devices.​‌ There is a real​​ need for new standards​​​‌ in lightweight cryptography, and​ the selected algorithms are​‌ expected to be widely​​ deployed within the Internet​​​‌ of Things, as well​ as on more constrained​‌ devices such as contactless​​ smart cards, or medical​​​‌ implants. The NIST received​ 56 submissions in February​‌ 2019. Team Capsule has​​ studied the security of​​​‌ some of these schemes.​

Monitoring Current Standards.

While​‌ we are very involved​​ in the design phase​​​‌ of new cryptographic standards,​ we also monitor the​‌ algorithms that are already​​ standardized. We look at​​​‌ some implementations of WPA3​ and we discovered a​‌ micro-architectural attack 56.​​ We also studied the​​​‌ privacy of the EME​ standard (Encrypted Media Extensions)​‌ for Digital Rights Managment​​ in browsers in 89​​​‌.

5 Social and​ environmental responsibility

5.1 Impact​‌ of research results

The​​ work 29 has been​​​‌ improved by Craig Gidney​ in a new evaluation​‌ for Shor's algorithm on​​ Google Quantum architecture 73​​​‌. It shows that​ less than 1400 logical​‌ qubits, or about 1​​ million physical qubits, are​​​‌ needed to factor RSA​ 2048 bits if a​‌ quantum computer is built.​​ This result has been​​​‌ presented in 2024 in​ QIP and in 2025​‌ in CRYPTO. André Schrottenloher​​ also gave a talk​​ at the Simons Institute​​​‌ (Berkeley) in Summer 2025,‌ Clémence Chevignard at Quantum‌​‌ Innovation 2025 in Nagoya,​​ and Pierre-Alain Fouque in​​​‌ the Tavares Lecture at‌ the Selected Area in‌​‌ Cryptography, SAC Conference, Toronto​​ 2025.

The same algorithm​​​‌ can be used to‌ attack the discrete logarithm‌​‌ problem in finite field,​​ such as safe prime​​​‌ field. In the specific‌ case of IKE (Internet‌​‌ Key Exchange), the key​​ exchange algorithm used in​​​‌ IPSEC, with short discrete‌ log, the attack only‌​‌ requires 300 logical qubits​​ to be implemented.

6​​​‌ Highlights of the year‌

  1. The paper 13,‌​‌ published at Eurocrypt 2025,​​ presents a blockcipher for​​​‌ encrypting program instructions in‌ order to avoid timing‌​‌ attacks on cryptographic implementations.​​
  2. The paper 15,​​​‌ published at CRYPTO 2025,‌ describes a dual attack‌​‌ on Kyber and reassesses​​ the security of the​​​‌ standard.

6.1 Awards

  1. The‌ paper 25, published‌​‌ at DSN 2025, has​​ received the best paper​​​‌ award.

7 Latest software‌ developments, platforms, open data‌​‌

7.1 Latest software developments​​

7.1.1 TNFS-alpha

  • Name:
    alpha​​​‌ for the Tower Number‌ Field Sieve algorithm
  • Keyword:‌​‌
    Cryptography
  • Functional Description:
    This​​ library implements a simulation​​​‌ tool for the tower‌ number field sieve algorithm‌​‌ computing discrete logarithms in​​ extension fields of small​​​‌ degree (tested up to‌ 54). The library contains‌​‌ an implementation of the​​ exact computation of alpha,​​​‌ the bias between the‌ expected smoothness of an‌​‌ integer and the expected​​ smoothness of a norm​​​‌ of an algebraic integer‌ in a number field‌​‌ made of two extensions.​​ The algorithm is a​​​‌ generalization to extensions of‌ the exact implementation of‌​‌ alpha in the software​​ CADO-NFS. The software contains​​​‌ an implementation of the‌ E-function of B. A.‌​‌ Murphy (Murphy's E) which​​ estimates the quality of​​​‌ the polynomial selection step‌ in TNFS through a‌​‌ simulation of the yield​​ of the relation collection​​​‌ in the TNFS algorithm.‌ Finally, it contains a‌​‌ database of pairing-friendly curve​​ seeds with the estimated​​​‌ level of security w.r.t.‌ a discrete logarithm computation‌​‌ in the corresponding finite​​ field.
  • News of the​​​‌ Year:
    In 2025, new‌ curves in the family‌​‌ Gasnier-Guillevic were added. As​​ part of the PEPR​​​‌ Cryptanalyse, new tables of‌ polynomials with automorphisms were‌​‌ included to the project​​ (degrees 6, 8, 9).​​​‌ Finally, new tables of‌ sparse polynomials of degrees‌​‌ 19 and 26 were​​ included.
  • URL:
  • Publications:​​​‌
  • Contact:​​
    Aurore Guillevic
  • Participant:
    Aurore​​​‌ Guillevic

7.1.2 Qarton

  • Keywords:‌
    Quantum programming, Cryptography, Post-quantum,‌​‌ Quantum cryptanalysis
  • Functional Description:​​
    Qarton is a python​​​‌ library to represent, analyze,‌ simulate and optimize medium‌​‌ and large-scale quantum circuits​​ at the logical level,​​​‌ with a focus on‌ circuits arising in cryptanalysis.‌​‌
  • Release Contributions:
    First version​​ of the library.
  • URL:​​​‌
  • Contact:
    Andre Schrottenloher‌

8 New results

8.1‌​‌ Secret-Key Cryptography

8.1.1 Proofs​​ of Security of Symmetric​​​‌ Constructions

Participants: André Schrottenloher‌.

Post-quantum Security of‌​‌ Key-Alternating Feistel Ciphers

 12​​

Since Kuwakado and Morii's​​​‌ work 81, 82‌ it is known that‌​‌ the classically secure 3-round​​​‌ Luby-Rackoff PRP and Even-Mansour​ cipher become insecure against​‌ an adversary equipped with​​ quantum query access. However,​​​‌ while this query model​ (the so-called Q2 model)​‌ has led to many​​ more attacks, it seems​​​‌ that restricting the adversary​ to classical query access​‌ prevents such breaks (the​​ so-called Q1 model). Indeed,​​​‌ at EUROCRYPT 2022, Alagic​ et al. 39 proved​‌ the Q1-security of the​​ Even-Mansour cipher.

In the​​​‌ paper 12, we​ focus on Feistel ciphers.​‌ More precisely, we consider​​ Key-Alternating Feistels built from​​​‌ random functions or permutations​ and random independent round​‌ keys. We borrow the​​ tools used by Alagic​​​‌ et al. and adapt​ them to this setting,​‌ showing that in the​​ Q1 setting:  the​​​‌ 3-round Key-Alternating Feistel, even​ when the round functions​‌ are the same random​​ oracle, is a pseudo-random​​​‌ permutation;  similarly the​ 4-round KAF is a​‌ strong pseudo-random permutation.

8.1.2​​ Symmetric Cryptanalysis of Primitives​​​‌ and Tools

Simplified Meet-in-the-middle​ Preimage Attacks on AES-based​‌ Hashing

7

Participants: Mathieu​​ Degré, Patrick Derbez​​​‌, André Schrottenloher.​

The meet-in-the-middle (MITM) attack​‌ is a powerful cryptanalytic​​ technique leveraging time-memory tradeoffs​​​‌ to break cryptographic primitives.​ Initially introduced for block​‌ cipher cryptanalysis, it has​​ since been extended to​​​‌ hash functions, particularly preimage​ attacks on AES-based compression​‌ functions. Over the years,​​ various enhancements such as​​​‌ superposition MITM 40 and​ bidirectional propagations have significantly​‌ improved MITM attacks, but​​ at the cost of​​​‌ increasing complexity of automated​ search models. In this​‌ work, we propose a​​ unified mixed integer linear​​​‌ programming (MILP) model designed​ to improve the search​‌ for optimal pre-image MITM​​ attacks against AES-based compression​​​‌ functions. Our model generalizes​ previous approaches by simplifying​‌ both the modeling and​​ the corresponding attack algorithm.​​​‌ In particular, it ensures​ that all identified attacks​‌ are valid. Our framework​​ not only recovers known​​​‌ attacks on AES and​ Whirlpool but also discovers​‌ new attacks with lower​​ memory complexities, and new​​​‌ quantum attacks.

We made​ the code of our​‌ model, and all applications​​ given in the paper,​​​‌ available on the Inria​ GitLab platform.

New​‌ Models for the Cryptanalysis​​ of ASCON.

6

Participants:​​​‌ Mathieu Degré, Patrick​ Derbez, André Schrottenloher​‌.

This paper focuses​​ on the cryptanalysis of​​​‌ the ASCON family using​ automatic tools. We analyze​‌ two different problems with​​ the goal to obtain​​​‌ new modelings, both simpler​ and less computationally heavy​‌ than previous works (all​​ our models require only​​​‌ a small amount of​ code and run on​‌ regular desktop computers).

The​​ first problem is the​​​‌ search for Meet-in-the-middle attacks​ on reduced-round ASCON-Hash. Starting​‌ from a previous MILP​​ modeling of Qin et​​​‌ al. (EUROCRYPT 2023), we​ rephrase the problem in​‌ SAT, which accelerates significantly​​ the solving time and​​​‌ removes the need for​ the “weak diffusion structure”​‌ heuristic. This allows us​​ to reduce the memory​​​‌ complexity of Qin et​ al.'s attacks and to​‌ prove some optimality results.​​

The second problem is​​​‌ the search for lower​ bounds on the probability​‌ of differential characteristics for​​ the ASCON permutation. We​​ introduce a lossy MILP​​​‌ encoding of the propagation‌ rules based on the‌​‌ Hamming weight, in order​​ to find quickly lower​​​‌ bounds which are comparable‌ to the state of‌​‌ the art. We find​​ a small improvement over​​​‌ the existing bound on‌ 7 rounds.

This paper‌​‌ is the full version,​​ published in Design, Codes​​​‌ and Cryptography 2025. A‌ previous version was presented‌​‌ at WCC 2024.​​

ChiLow and ChiChi: New​​​‌ Constructions for Code Encryption.‌

13

Participants: Patrick Derbez‌​‌.

We study the​​ problem of embedded code​​​‌ encryption, i.e., encryption for‌ binary software code for‌​‌ a secure microcontroller that​​ is stored in an​​​‌ insecure external memory. As‌ every single instruction must‌​‌ be decrypted before it​​ can be executed, this​​​‌ scenario requires an extremely‌ low latency decryption. We‌​‌ present a formal treatment​​ of embedded code encryption​​​‌ security definitions, propose three‌ constructions, namely ACE1, ACE2‌​‌ and ACE3, and analyze​​ their security. Further, we​​​‌ present ChiLow, a family‌ of tweakable block ciphers‌​‌ and a related PRF​​ specifically designed for embedded​​​‌ code encryption. At the‌ core of ChiLow, there‌​‌ is ChiChi, a new​​ family of non-linear layers​​​‌ of even dimension based‌ on the well-known χ‌​‌ function. Our fully unrolled​​ hardware implementation of ChiLow,​​​‌ using the Nangate 15nm‌ Open Cell Library, achieves‌​‌ a decryption latency of​​ less than 280 picoseconds.​​​‌

Improved Cryptanalysis of GIFT-64.‌

8

Participants: Patrick Derbez‌​‌, Baptiste Germon.​​

In this paper, we​​​‌ propose new differential attacks‌ against the block cipher‌​‌ GIFT-64. First we demonstrate​​ how the parallel matching​​​‌ algorithm proposed by Naya-Plasencia‌ at CRYPTO’11 as an‌​‌ advanced list-merging algorithm can​​ be leveraged to enhance​​​‌ differential attacks, overcoming a‌ previously assumed bottleneck. By‌​‌ reducing the complexity of​​ the pairs generation process​​​‌ whenever a non-linear filter‌ is available, this approach‌​‌ enabled us to mount​​ a new differential attack​​​‌ against 25-round GIFT-64 in‌ the related-key setting.Then we‌​‌ use the differential Meet-in-the-Middle​​ cryptanalysis technique introduced by​​​‌ Boura et al. at‌ CRYPTO’23 to improve the‌​‌ differential attacks recently proposed​​ by Chang et al.​​​‌ at CT-RSA’25, leading to‌ the best known attacks‌​‌ against GIFT-64 in the​​ single-key setting, both in​​​‌ terms of number of‌ rounds and of complexity.‌​‌

Extending the Quasidifferential Framework:​​ From Fixed-Key to Expected​​​‌ Differential Probability.

5

Participants:‌ Patrick Derbez, Baptiste‌​‌ Germon.

Beyne and​​ Rijmen proposed in 2022​​​‌ a systematic and generic‌ framework to study the‌​‌ fixed-key probability of differential​​ characteristics. One of the​​​‌ main challenges for implementing‌ this framework is the‌​‌ ability to efficiently handle​​ very large quasidifferential transition​​​‌ matrices (QDTMs) for big‌ (e.g. 8-bit) S-boxes. Our‌​‌ first contribution is a​​ new MILP model capable​​​‌ of efficiently representing such‌ matrices, by exploiting the‌​‌ inherent block structure of​​ these objects. We then​​​‌ propose two extensions to‌ the original framework. First,‌​‌ we demonstrate how to​​ adapt the framework to​​​‌ the related-key setting. Next,‌ we present a novel‌​‌ approach to compute the​​ average expected probability of​​​‌ a differential characteristic that‌ takes the key schedule‌​‌ into account. This method,​​​‌ applicable to both linear​ and non-linear key schedules,​‌ works in both the​​ single-key and related-key settings.​​​‌ Furthermore, it provides a​ faster way to verify​‌ the validity of characteristics​​ compared to computing the​​​‌ fixed-key probability. Using these​ extensions and our MILP​‌ model, we analyze various​​ (related-key) differential characteristics from​​​‌ the literature. First, we​ prove the validity of​‌ several optimal related-key differential​​ characteristics of AES. Next,​​​‌ we show that this​ approach permits to obtain​‌ more precise results than​​ methods relying on key​​​‌ constraints for SKINNY. Finally,​ we examine the validity​‌ of a differential distinguisher​​ used in two differential​​​‌ meet-in-the-middle attacks on SKINNY-128,​ demonstrating that its probability​‌ is significantly higher than​​ initially estimated.

Minimalist Model​​​‌ for Impossible Differentials.

20​

Participants: Patrick Derbez,​‌ Marie Euler.

This​​ paper introduces a new​​​‌ MILP modeling to find​ impossible differential (ID) distinguishers​‌ and attacks. Standard models​​ for ID are negative​​​‌ models, in the sense​ that a differential is​‌ impossible if and only​​ if the model has​​​‌ no solution. Our new​ modelling technique focuses on​‌ probable ID, differentials that​​ are probably impossible. While​​​‌ this might lead to​ false positives, the main​‌ advantage is that searching​​ for such probable ID​​​‌ can be achieved through​ a positive model. This​‌ facilitates the search for​​ the best impossible differential​​​‌ attacks without first exhausting​ all possible ID distinguishers​‌ on a target. We​​ also propose to simplify​​​‌ the modelling by only​ considering two possible states​‌ for internal cells: inactive​​ and unknown. In this​​​‌ case there are no​ longer direct contradictions but​‌ only indirect ones, assuming​​ that it is impossible​​​‌ that all cells are​ inactive.

With these two​‌ simple ideas, we are​​ able to retrieve the​​​‌ longest impossible differentials distinguishers​ on MIDORI, SKINNY, PRESENT,​‌ SIMON, Simeck and SPECK.​​ Furthermore, as the model​​​‌ looking for candidates is​ based on satisfiability, it​‌ can be incorporated in​​ a larger model which​​​‌ looks directly for the​ best attacks in order​‌ to enumerate the distinguishers​​ in the order of​​​‌ the complexity of the​ associated attacks, which we​‌ did for the AES,​​ ARADI, SIMON and SKINNY.​​​‌

SPEEDY: Caught at Last.​

14

Participants: Patrick Derbez​‌, Baptiste Germon.​​

SPEEDY is a family​​​‌ of ultra-low-latency block ciphers​ designed by Leander et​‌ al. in 2021. In​​ 2023, Boura et al.​​​‌ proposed a differential attack​ on the full 7-round​‌ variant, SPEEDY-7-192. However, shortly​​ thereafter, Beyne and Neyt​​​‌ demonstrated that this attack​ was invalid, as the​‌ dominant differential characteristic it​​ relied upon had probability​​​‌ zero. A similar issue​ affects another differential attack​‌ proposed the same year​​ by Wang et al.,​​​‌ which also targets SPEEDY-7-192​ and suffers from the​‌ same flaw. As a​​ result, the question of​​​‌ finding a valid attack​ on this cipher remained​‌ an open problem. In​​ this work, we resolve​​​‌ this problem by presenting​ the first valid differential​‌ attack on SPEEDY-7-192. We​​ verify the validity of​​​‌ our distinguisher using the​ quasidifferential framework. Moreover, our​‌ search for the differential​​ distinguisher is significantly more​​ rigorous than in previous​​​‌ works: starting from a‌ pool of one-round trails,‌​‌ our method explores a​​ larger portion of the​​​‌ search space. We also‌ fully exploit probabilistic extensions‌​‌ of the distinguisher to​​ identify optimal parameters for​​​‌ the key recovery step.‌ Our best attack on‌​‌ SPEEDY-7-192 is a chosen-ciphertext​​ attack with data and​​​‌ time complexity 2174‌.53. In‌​‌ addition, we present differential​​ attacks on 4-round SPEEDY-5-192​​​‌ and 5-round SPEEDY-6-192, which‌ currently represent the b‌​‌ est known attacks against​​ these smaller variants.

8.1.3​​​‌ Quantum Cryptanalysis

Faster Quantum‌ Algorithms for MQ2 and‌​‌ Applications.

9

Participants: Pierre-Alain​​ Fouque, André Schrottenloher​​​‌.

In this work,‌ we study quantum algorithms‌​‌ for multivariate quadratic Boolean​​ equation systems by focusing​​​‌ on their precise gate‌ count. While better asymptotic‌​‌ algorithms are known, currently​​ gate counts were only​​​‌ computed for exhaustive search‌ and a variant of‌​‌ Grover's search using preprocessing​​ 92. This limits​​​‌ the applicability of Boolean‌ equation solving to cryptanalysis,‌​‌ which considers relatively small​​ numbers of variables (from​​​‌ 40 to 200) and‌ is concerned with the‌​‌ exact complexity of the​​ solver.

In this paper,​​​‌ we introduce two new‌ quantum algorithms: an optimized‌​‌ quantum exhaustive search, which​​ amortizes the cost of​​​‌ polynomial evaluation, and a‌ simple linearization strategy based‌​‌ on 53. We​​ apply these algorithms to​​​‌ the cryptanalysis of the‌ block ciphers LowMC and‌​‌ RAIN in the single-data​​ setting, which is important​​​‌ in the context of‌ post-quantum digital signatures based‌​‌ on MPC-in-the-head. This allows​​ us to adapt existing​​​‌ classical attacks into the‌ first quantum cryptanalysis results‌​‌ on these ciphers.

We​​ implemented the main building​​​‌ blocks of the circuits‌ presented in this paper,‌​‌ and published this implementation​​ on the Inria Gitlab​​​‌ platform.

Improved Quantum‌ Linear Attacks and Application‌​‌ to CAST.

4

Participants:​​ André Schrottenloher.

This​​​‌ paper studies quantum linear‌ key-recovery attacks on block‌​‌ ciphers. The first such​​ attacks were last-rounds attacks​​​‌ proposed by Kaplan et‌ al. 77, which‌​‌ combine a linear distinguisher​​ with a guess of​​​‌ a subkey. More recently,‌ the framework which we‌​‌ introduced in 95 uses​​ a quantum convolution algorithm​​​‌ to compute a so-called‌ correlation state, which‌​‌ is a superposition of​​ subkey candidates where the​​​‌ amplitudes are linear correlations.‌ The main limitation of‌​‌ this approach is that,​​ while the good subkey​​​‌ has the highest correlation,‌ it is not marked‌​‌ in the state, and​​ cannot be found immediately​​​‌ by quantum search.

In‌ this paper, we combine‌​‌ the correlation state with​​ a distinguisher, which recognizes​​​‌ the good subkey. From‌ here, we can use‌​‌ quantum search to recover​​ this key. We apply​​​‌ this idea to Feistel‌ ciphers, notably the CAST-128‌​‌ and CAST-256 ciphers, using​​ two kinds of distinguishers:​​​‌ quantum distinguishers based on‌ Simon's algorithm 81 and‌​‌ linear distinguishers. The resulting​​ attacks outperform the previous​​​‌ quantum attacks.

8.2 Public-key‌ cryptography

8.2.1 Lattices

Assessing‌​‌ the Impact of a​​ Variant of MATZOV's Dual​​​‌ Attack on Kyber

15‌

Participants: Charles Meyer-Hilfiger,‌​‌ Yixin Shen.

The​​​‌ dual attacks on the​ Learning With Errors problem​‌ are currently a subject​​ of controversy. In particular,​​​‌ the results of 85​, which claim to​‌ significantly lower the security​​ level of Kyber 96​​​‌, a lattice-based cryptosystem​ currently being standardized by​‌ NIST, are not widely​​ accepted. The analysis behind​​​‌ their attack depends on​ a series of assumptions​‌ that, in certain scenarios,​​ have been shown to​​​‌ contradict established theorems or​ well-tested heuristics 37.​‌

In this paper, we​​ introduce a new dual​​​‌ lattice attack on LWE,​ drawing from ideas in​‌ coding theory. Our approach​​ revisits the dual attack​​​‌ proposed by 85,​ replacing modulus switching with​‌ an efficient decoding algorithm.​​ This decoding is achieved​​​‌ by generalizing polar codes​ over Zq,​‌ and we confirm their​​ strong distortion properties through​​​‌ benchmarks. This modification enables​ a reduction from small-LWE​‌ to plain-LWE, with a​​ notable decrease in the​​​‌ secret dimension. Additionally, we​ replace the enumeration step​‌ in the attack by​​ assuming the secret is​​​‌ zero for the portion​ being enumerated, iterating this​‌ assumption over various choices​​ for the enumeration part.​​​‌

We make an analysis​ of our attack without​‌ using the flawed independence​​ assumptions used in 85​​​‌ and we fully back​ up our analysis with​‌ experimental evidence.

Lastly, we​​ assess the complexity of​​​‌ our attack on Kyber;​ showing that the security​‌ levels for Kyber-512/768/1024 are​​ 3.5/11.9/12.3 bits below the​​​‌ NIST requirements (143/207/272 bits)​ in the same nearest-neighbor​‌ cost model as in​​ 96, 85.​​​‌ All in all the​ cost of our attack​‌ matches and even slightly​​ beat in some cases​​​‌ the complexities originally claimed​ by the attack of​‌ 85.

Discrete gaussian​​ sampling for BKZ-reduced basis​​​‌

28

Participants: Yixin Shen​.

Discrete Gaussian sampling​‌ on lattices is a​​ fundamental problem in lattice-based​​​‌ cryptography. In this paper​ 28, we revisit​‌ the Markov chain Monte​​ Carlo (MCMC)-based Metropolis-Hastings-Klein (MHK)​​​‌ algorithm proposed by Wang​ and Ling and study​‌ its complexity under the​​ Geometric Series Assuption (GSA)​​​‌ when the given basis​ is BKZ-reduced. We give​‌ experimental evidence that the​​ GSA is accurate in​​​‌ this context, and we​ give a very simple​‌ approximate formula for the​​ complexity of the sampler​​​‌ that is accurate over​ a large range of​‌ parameters and easily computable.​​ We apply our results​​​‌ to the dual attack​ on LWE of 91​‌ and significantly improve the​​ complexity estimates of the​​​‌ attack. Finally, we provide​ some results of independent​‌ interest on the Gaussian​​ mass of a random​​​‌ q-ary lattices.

A​ reduction from Hawk to​‌ the principal ideal problem​​ in a quaternion algebra​​​‌

19

Participants: Clémence Chevignard​.

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).

Ideally​​​‌ HAWKward: How Not to‌ Break Module-LIP

30

Participants:‌​‌ Clémence Chevignard.

The​​ module-Lattice Isomorphism Problem (module-LIP)​​​‌ was introduced by Ducas‌ et al. in 67‌​‌, and used within​​ the signature scheme and​​​‌ NIST candidate HAWK. In‌ 88, Mureau et‌​‌ al. pointed out that​​ over certain number fields​​​‌ F , the problem‌ can be reduced to‌​‌ enumerating solutions of x​​2+y2​​​‌=q where q‌𝒪F is‌​‌ given and x,​​yOF​​​‌ are the unknowns). Moreover‌ one can always reduce‌​‌ to a similar equation​​ which has only few​​​‌ solutions. This key insight‌ led to a heuristic‌​‌ polynomial-time algorithm for solving​​ module-LIP on those specific​​​‌ instances. Yet this result‌ doesn't threaten HAWK for‌​‌ which the problem can​​ be reduced to enumerating​​​‌ solutions of x2‌+y2+‌​‌z2+t​​2=q (where​​​‌ q𝒪F‌ is given and x‌​‌,y,z​​,t𝒪​​​‌F are the unknowns.‌ We show that, in‌​‌ all likelihood, solving this​​ equation requires the enumeration​​​‌ of a too large‌ set to be feasible,‌​‌ thereby making irrelevant a​​ straightforward adaptation of the​​​‌ approach in 88.‌

Share the MAYO: Thresholdizing-MAYO‌​‌

16

Participants: Guilhem Niot​​.

Threshold cryptography is​​​‌ a growing field that‌ allows multiple parties to‌​‌ jointly perform cryptographic operations​​ without exposing their individual​​​‌ secret shares. In this‌ paper, we present the‌​‌ first comprehensive study on​​ thresholdizing practical OV-based signature​​​‌ schemes, specifically focusing on‌ MAYO and UOV, which‌​‌ are candidates in the​​ NIST process for standardization​​​‌ of additional digital signature‌ schemes. Our approach begins‌​‌ by addressing the challenges​​ associated with thresholdizing algorithms​​​‌ that sample solutions to‌ linear equation systems of‌​‌ the form 𝐀x​​=y, which​​​‌ are fundamental to OV-based‌ signature schemes. Previous attempts‌​‌ have introduced levels of​​ leakage that we deem​​​‌ insecure. We propose a‌ novel minimum-leakage solution and‌​‌ assess its practicality. Furthermore,​​ we explore the thresholdization​​​‌ of the entire functionality‌ of these signature schemes,‌​‌ demonstrating their unique applications​​ in networks and cryptographic​​​‌ protocols.

Finally! A Compact‌ Lattice-Based Threshold Signature

27‌​‌

Participants: Guilhem Niot.​​

Threshold signatures split trust​​​‌ among parties, requiring T‌ of N to sign.‌​‌ While common in pre-quantum​​ cryptography, post-quantum threshold schemes​​​‌ remain heavy, with signature‌ sizes an order of‌​‌ magnitude larger than standard​​ PQ signatures.

We propose​​​‌ a novel, highly efficient‌ threshold signature scheme with‌​‌ sizes close to a​​​‌ standard ML-DSA signature for​ T8.​‌ Our construction relies on​​ well-studied assumptions (MLWE and​​​‌ SelfTargetMSIS) and avoids heavy​ machinery, essentially running T​‌ parallel Dilithium executions. Despite​​ its simplicity, achieving this​​​‌ required overcoming technical hurdles​ like small share distribution​‌ and rejecting transcript simulation,​​ delivering a previously out-of-reach​​​‌ efficiency.

Efficient Threshold ML-DSA​

17

Participants: Guilhem Niot​‌.

In this paper,​​ we present the first​​​‌ threshold signature scheme fully​ compatible with the NIST-standardized​‌ ML-DSA. While existing lattice-based​​ threshold solutions either lack​​​‌ practicality or standard compliance,​ our approach supports secure​‌ and efficient production of​​ ML-DSA-compatible signatures for small​​​‌ groups, requiring under 1​ MB of communication per​‌ party for up to​​ 6 signers. We leverage​​​‌ advanced short secret sharing​ and optimized rejection sampling​‌ to achieve a practical​​ balance between communication efficiency​​​‌ and number of rounds​ required to output a​‌ valid signature. We implement​​ our construction in Go​​​‌ and provide benchmarks across​ LAN/WAN settings to demonstrate​‌ practical deployability for applications​​ such as cryptocurrency wallets,​​​‌ threshold TLS, and Tor​ directory authorities.

Unmasking TRaccoon:​‌ A Lattice-Based Threshold Signature​​ with An Efficient Identifiable​​​‌ Abort Protocol

26

Participants:​ Guilhem Niot.

TRaccoon​‌ is an efficient 3-round​​ lattice-based threshold signature, recently​​​‌ introduced by del Pino​ et al. 90.​‌ While the design resembles​​ the classical threshold Schnorr​​​‌ signature Sparkle, it lacks​ a means to identify​‌ malicious behavior—-a property of​​ interest in practice. This​​​‌ limitation stems from TRaccoon's​ use of masking to​‌ resist lattice-specific attacks, which​​ blinds partial signatures with​​​‌ one-time additive masks. del​ Pino et al. left​‌ the addition of an​​ identification mechanism as an​​​‌ open problem.

In this​ work, we propose TRaccoon-IA,​‌ which extends TRaccoon with​​ an efficient identifiable abort​​​‌ protocol to identify malicious​ signers when the protocol​‌ fails. This simple add-on​​ preserves the original design​​​‌ and incurs an added​ communication cost of 60​‌+6.4​​|T| KB​​​‌ only in the event​ of a failure. Additionally,​‌ we provide the first​​ formal security analysis of​​​‌ a zero-knowledge variant of​ LaBRADOR and introduce a​‌ new game-based definition for​​ interactive identifiable abort protocols,​​​‌ extending standard unforgeability definitions.​

Practical Deniable Post-Quantum X3DH:​‌ A Lightweight Split-KEM for​​ K-Waay

24

Participants: Guilhem​​​‌ Niot.

The Signal​ Protocol faces the challenge​‌ of migrating to a​​ post-quantum world while preserving​​​‌ critical properties such as​ asynchrony and deniability. While​‌ PQXDH grants post-quantum confidentiality,​​ full migration of the​​​‌ X3DH handshake remains elusive.​ K-Waay 61 offers a​‌ path via split KEMs​​ but suffers from size​​​‌ limitations compared to ring​ signature-based approaches.

This work​‌ introduces Sparrow-KEM and Sym-Sparrow-KEM,​​ novel asymmetric and symmetric​​​‌ split KEMs designed to​ optimize K-Waay. Leveraging the​‌ MLWE assumption, we reduce​​ communication by 5.1×​​​‌ and improve speed by​ 40× over prior​‌ split KEMs. Sym-Sparrow-KEM is​​ the first symmetric split-KEM​​​‌ to offer deniability along​ with strong implicit authentication​‌ properties (IND-1KCA, IND-1BatchCCA). Our​​ results demonstrate the feasibility​​​‌ of a compact, deniable​ post-quantum X3DH based on​‌ split KEMs.

8.2.2 Elliptic​​ curves and isogenies

Participants:​​ Aurore Guillevic.

An​​​‌ algebraic point of view‌ on the generation of‌​‌ pairing-friendly curves

The paper​​ 11 with Jean Gasnier​​​‌ from the CANARI Team‌ (Bordeaux) is the achievement‌​‌ of Jean Gasnier's Masters​​ internship in 2022 co-advised​​​‌ in Bordeaux by Jean-Marc‌ Couveignes and remotely from‌​‌ Denmark by Aurore Guillevic,​​ and Gasnier's PhD thesis​​​‌ defended in Bordeaux in‌ July 2025. It aims‌​‌ to generalize The Kachisa–Schaefer–Scott​​ technique to find new​​​‌ parameterized families of pairing-friendly‌ curves. The method allowed‌​‌ to obtain new curves​​ for interesting embedding degrees,​​​‌ such as k=‌20. It comes‌​‌ with two implementations, one​​ written by Jean Gasnier​​​‌ to obtain new curve‌ families (see Subfield Method‌​‌ Gitlab Project), the​​ other one to implement​​​‌ pairings on the new‌ curves, see Pairings on‌​‌ Gasnier–Guillevic Curves Gitlab Project​​. Finally the paper​​​‌ is published in the‌ journal SIAGA.

Participants: Pierrick‌​‌ Dartois.

qt-Pegasis: Simpler​​ and Faster Effective Class​​​‌ Group Actions

The paper‌ 36 by Pierrick Dartois,‌​‌ Jonathan Komada Eriksen, Riccardo​​ Invernizzi, and Frederik Vercauteren,​​​‌ revisits the recent Pegasis‌ algorithm that computes an‌​‌ effective group action of​​ the class group of​​​‌ any imaginary quadratic order‌ R on a set‌​‌ of supersingular elliptic curves​​ primitively oriented by R​​​‌. Although Pegasis was‌ the first algorithm showing‌​‌ the practicality of computing​​ unrestricted class group actions​​​‌ at higher security levels,‌ it is complicated and‌​‌ prone to failures, which​​ leads to many rerandomizations.​​​‌

In this work, we‌ present a new algorithm,‌​‌ qt-Pegasis, which is much​​ simpler, but at the​​​‌ same time faster and‌ removes the need for‌​‌ rerandomization of the ideal​​ we want to act​​​‌ with, since it never‌ fails. It leverages the‌​‌ main technique of the​​ recent Qlapoti approach. However,​​​‌ Qlapoti solves a norm‌ equation in a quaternion‌​‌ algebra, which corresponds to​​ the full endomorphism ring​​​‌ of a supersingular elliptic‌ curve. We show that‌​‌ the algorithm still applies​​ in the quadratic setting,​​​‌ by embedding the quadratic‌ ideal into a quaternion‌​‌ ideal using a technique​​ similar to the one​​​‌ applied in KLaPoTi. This‌ way, we can reinterpret‌​‌ the output of Qlapoti​​ as four equivalent quadratic​​​‌ ideals, instead of two‌ equivalent quaternion ideals. We‌​‌ then show how to​​ construct a Clapoti-like diagram​​​‌ in dimension 2, which‌ embeds the action of‌​‌ the ideal in a​​ 4-dimensional isogeny. We implemented​​​‌ our qt-Pegasis algorithm in‌ SageMath for the CSURF‌​‌ group action, and we​​ achieve a speedup over​​​‌ Pegasis of 1.8×‌ for the 500-bit parameters‌​‌ and 2.6× for​​ the 4000-bit parameters.

8.2.3​​​‌ Quantum Cryptanalysis

Improved Classical‌ and Quantum Algorithms for‌​‌ the Shortest Vector Problem​​ via Bounded Distance Decoding​​​‌

3

Participants: Yixin Shen‌.

The most important‌​‌ computational problem on lattices​​ is the Shortest Vector​​​‌ Problem (SVP). In this‌ paper, we present new‌​‌ algorithms that improve the​​ state-of-the-art for provable classical/quantum​​​‌ algorithms for SVP. We‌ present the following results.‌​‌

  1. A new algorithm for​​ SVP that provides a​​​‌ smooth tradeoff between time‌ complexity and memory requirement.‌​‌ For any positive integer​​​‌ 4q≤​n, our algorithm​‌ takes q13n​​+o(n​​​‌) time and requires​ poly​‌(n)·​​q16n/​​​‌q2 memory. This​ tradeoff which ranges from​‌ enumeration (q=​​n) to sieving​​​‌ (q constant), is​ a consequence of a​‌ new time-memory tradeoff for​​ Discrete Gaussian sampling above​​​‌ the smoothing parameter.
  2. A​ quantum algorithm for SVP​‌ that runs in time​​ 20.950​​​‌n+o(​n) and requires​‌ 20.5​​n+o(​​​‌n) classical memory​ and poly(n​‌) qubits. In Quantum​​ Random Access Memory (QRAM)​​​‌ model this algorithm takes​ only 20.​‌835n+o​​(n) time​​​‌ and requires a QRAM​ of size 20​‌.293n+​​o(n)​​​‌, poly(n​) qubits and 2​‌0.5n​​ classical space. This improves​​​‌ over the previously fastest​ classical (which is also​‌ the fastest quantum) algorithm​​ due to 38 that​​​‌ has a time and​ space complexity 2n​‌+o(n​​).
  3. A classical​​​‌ algorithm for SVP that​ runs in time 2​‌1.669n​​+o(n​​​‌) time and 2​0.5n​‌+o(n​​) space. This improves​​​‌ over an algorithm of​ 58 that has the​‌ same space complexity.

The​​ time complexity of our​​​‌ classical and quantum algorithms​ are obtained using a​‌ known upper bound on​​ a quantity related to​​​‌ the lattice kissing number​ which is 20​‌.402n.​​ We conjecture that for​​​‌ most lattices this quantity​ is a 2o​‌(n).​​ Assuming that this is​​​‌ the case, our classical​ algorithm runs in time​‌ 21.292​​n+o(​​​‌n), our​ quantum algorithm runs in​‌ time 20.​​750n+o​​​‌(n) and​ our quantum algorithm in​‌ QRAM model runs in​​ time 20.​​​‌667n+o​(n).​‌ As a direct application​​ of our result, using​​​‌ the reduction in 68​, we obtain a​‌ provable quantum algorithm for​​ the Lattice Isomorphism Problem​​​‌ in the case of​ the trivial lattice Z​‌n (ZLIP) that runs​​ in time 20​​​‌.417n+​o(n)​‌. Our algorithm requires​​ a QRAM of size​​​‌ 20.147​n+o(​‌n), poly​​(n) qubits​​​‌ and 20.​25n classical space.​‌

A Tight Quantum Algorithm​​ for Multiple Collision Search.​​​‌

34

Participants: André Schrottenloher​, Yixin Shen.​‌

Searching for collisions in​​ random functions is a​​​‌ fundamental computational problem, with​ many applications in symmetric​‌ and asymmetric cryptanalysis. When​​ one searches for a​​​‌ single collision, the known​ quantum algorithms match the​‌ query lower bound. This​​ is not the case​​ for the problem of​​​‌ finding multiple collisions, despite‌ its regular appearance as‌​‌ a sub-component in sieving-type​​ algorithms.

At EUROCRYPT 2019,​​​‌ Liu and Zhandry gave‌ a query lower bound‌​‌ Ω(2m​​/3+2​​​‌k/3)‌ for finding 2k‌​‌ collisions in a random​​ function with m-bit​​​‌ output. In a previous‌ paper at EUROCRYPT 2023‌​‌ 49, we gave​​ a quantum algorithm matching​​​‌ this bound for a‌ large range of m‌​‌ and k, but​​ not all admissible values.​​​‌ This quantum algorithm was‌ based on the MNRS‌​‌ quantum walk framework, with​​ the novelty that walks​​​‌ could be chained by‌ reusing the state after‌​‌ outputting a collision.

In​​ this paper, we give​​​‌ a new algorithm that‌ tackles the remaining non-optimal‌​‌ range, closing the problem.​​ Our algorithm is tight​​​‌ (up to a polynomial‌ factor) in queries, and‌​‌ also in time under​​ a quantum RAM assumption.​​​‌ The idea is to‌ extend the chained walk‌​‌ to a regime in​​ which several collisions are​​​‌ returned at each step,‌ and the “walks” themselves‌​‌ contain a single diffusion​​ step.

Reducing the Number​​​‌ of Qubits in Quantum‌ Factoring.

18

Participants: Clémence‌​‌ Chevignard, Pierre-Alain Fouque​​, André Schrottenloher.​​​‌

This paper focuses on‌ the optimization of the‌​‌ number of logical qubits​​ in quantum algorithms for​​​‌ factoring and computing discrete‌ logarithms in N‌​‌*. These algorithms​​ contain an exponentiation circuit​​​‌ modulo N, which‌ is responsible for most‌​‌ of their cost, both​​ in qubits and operations.​​​‌

In this paper, we‌ show that using only‌​‌ o(logN​​) work qubits, one​​​‌ can obtain the least‌ significant bits of the‌​‌ modular exponentiation output. We​​ combine this result with​​​‌ May and Schlieper's truncation‌ technique (ToSC 2022) and‌​‌ the Ekerå-Håstad variant of​​ Shor's algorithm (PQCrypto 2017)​​​‌ to solve the discrete‌ logarithm problem in ℤ‌​‌N* using only​​ d+o(​​​‌logN) qubits,‌ where d is the‌​‌ bit-size of the logarithm.​​ Consequently we can factor​​​‌ n-bit RSA moduli‌ using n/2‌​‌+o(n​​) qubits, while current​​​‌ envisioned implementations require about‌ 2n qubits.

Our‌​‌ algorithm uses a Residue​​ Number System and succeeds​​​‌ with a parametrizable probability.‌ Being completely classical, we‌​‌ have implemented and tested​​ it. For RSA factorization,​​​‌ we can reach a‌ gate count 𝒪(‌​‌n3) for​​ a depth 𝒪(​​​‌n2log3‌n), which‌​‌ then has to be​​ multiplied by 𝒪(​​​‌logn) (the‌ number of measurement results‌​‌ required by Ekerå-Håstad). To​​ factor an RSA-2048 instance,​​​‌ we estimate that 1730‌ logical qubits and 2‌​‌36 Toffoli gates will​​ suffice for a single​​​‌ run, and the algorithm‌ needs on average 40‌​‌ runs. To solve a​​ discrete logarithm instance of​​​‌ 224 bits (112-bit classical‌ security) in a safe-prime‌​‌ group of 2048 bits,​​ we estimate that 684​​​‌ logical qubits would suffice,‌ and 20 runs with‌​‌ 232 Toffoli gates​​​‌ each.

Our estimations are​ supported by a full​‌ implementation available on the​​ Inria Gitlab platform.​​​‌ This work was presented​ as a plenary talk​‌ in the QIP 2025​​ conference 29 and published​​​‌ in the proceedings of​ CRYPTO 2025 18.​‌ After our work was​​ initially made public in​​​‌ 2024, Gidney 73 performed​ an extensive estimate of​‌ physical estimates for breaking​​ RSA keys, combining an​​​‌ optimized version of our​ algorithm with up-to-date techniques​‌ in error correction. His​​ new estimate of a​​​‌ million physical qubits (down​ from 20 million in​‌ an earlier work 72​​) received significant attention.​​​‌

8.2.4 Protocols

Comprehensive Deniability​ Analysis of Signal Handshake​‌ Protocols: X3DH, PQXDH to​​ Fully Post-Quantum with Deniable​​​‌ Ring Signatures

31

Participants:​ Guilhem Niot.

The​‌ Signal protocol relies on​​ a handshake (formerly X3DH,​​​‌ now PQXDH) to set​ up secure conversations, valuing​‌ deniability so users can​​ deny participation. Prior analyses​​​‌ use varying, ad-hoc models​ that obscure guarantees and​‌ prevent comparison.

Building on​​ the abstraction by Hashimoto​​​‌ et al. 74,​ we present a unified​‌ framework for analyzing Signal​​ handshake deniability. We examine​​​‌ X3DH and PQXDH, clarifying​ PQXDH's deniability against harvest-now–judge-later​‌ quantum adversaries. We also​​ analyze post-quantum alternatives like​​​‌ RingXKEM that use ring​ signatures. By introducing a​‌ deniability metric inspired by​​ differential privacy, we offer​​​‌ relaxed, pragmatic guarantees. This​ metric further allows us​‌ to define deniable ring​​ signatures (a relaxation of​​​‌ anonymity), enabling efficient constructions​ from the NIST standard​‌ Falcon and the candidate​​ for standardization MAYO, which​​​‌ are deniable despite not​ being fully anonymous.

Revisiting​‌ PQ WireGuard: A Comprehensive​​ Security Analysis With a​​​‌ New Design Using Reinforced​ KEMs

22

Participants: Guilhem​‌ Niot.

WireGuard is​​ a high-performance VPN based​​​‌ on the Noise protocol.​ A recent post-quantum (PQ)​‌ variant was proposed by​​ Hülsing et al. 75​​​‌, however since Wireguard​ requires the handshake message​‌ to fit in one​​ UDP packet of size​​​‌ roughly 1200 B, they​ rely on Classic McEliece,​‌ whose large public keys​​ significantly increase server memory​​​‌ requirements and complicates kernel-level​ deployment.

In this work,​‌ we revisit PQ WireGuard​​ to improve its design,​​​‌ security, and efficiency. We​ address binding issues in​‌ PQ KEMs and prove​​ security in a new​​​‌ computational model. We introduce​ `reinforced KEM' (RKEM) and​‌ a construction named `Rebar'​​ to compress ML-KEM-like ciphertexts.​​​‌ This enables a PQ​ WireGuard protocol where the​‌ server avoids storing large​​ keys, reducing public key​​​‌ memory usage by 190​ to 390×.​‌

Subversion-resilient Key-exchange in the​​ Post-quantum World

21

Participants:​​​‌ Pierre-Alain Fouque, Guilhem​ Niot.

Subversion-resilient Authenticated​‌ Key-Exchange (AKE) ensures security​​ even when parts of​​​‌ the protocol implementation are​ tampered with. One way​‌ to achieve AKE is​​ by using Reverse Firewalls​​​‌ (RFs) to restore security.​

In this work, we​‌ extend RF-based subversion resilience​​ in security definitions, constructions,​​​‌ and formal verification. First,​ we introduce a useful​‌ relaxation of the notion​​ of security in subversion-resilient​​​‌ AKE with RFs: the​ goal is no longer​‌ to prevent all exfiltration,​​ but rather to restore​​ to the AKE protocol​​​‌ a property lost upon‌ subversion. We focus specifically‌​‌ on authenticating and (key-)securing​​ RFs, and consider a​​​‌ spectrum of compromises, designing‌ a framework in which‌​‌ adversaries can tamper with​​ some components of the​​​‌ implementation but perhaps not‌ others. Aiming for post-quantum‌​‌ security, we define `re-randomizable​​ Key Encapsulation Mechanisms', providing​​​‌ instantiations based on classical‌ Diffie-Hellman and Kyber. Finally,‌​‌ we establish foundations for​​ the formal verification of​​​‌ RF-based protocols, proving our‌ construction secure using the‌​‌ CryptoVerif prover, in addition​​ to computational-security proofs in​​​‌ usual Bellare-Rogaway methodology.

8.3‌ Side-Channel Attacks

Avengers assemble!‌​‌ Supervised learning meets lattice​​ reduction

Participants: Pierre-Alain Fouque​​​‌, Damien Marion,‌ Quyen Nguyen, Alexandre‌​‌ Wallet.

10

In​​ this work, we attack​​​‌ Kyber's key-generation algorithm using‌ power analysis and lattice‌​‌ reduction. More specifically, we​​ target the Centered Binomial​​​‌ Distribution (CBD) sampler which‌ generates the secret data‌​‌ of the underlying Learning​​ With Error (LWE) instance.​​​‌ From a side-channel perspective,‌ our attack uses a‌​‌ single trace, leveraging classifiers​​ developed through supervised learning.​​​‌ We assess the block-size‌ in lattice reduction that‌​‌ would complete the key​​ recovery, providing a fine-grained​​​‌ trade-offs between the correctly‌ guessed proportion and the‌​‌ block-size, based on standard​​ estimates. Finally, we conducted​​​‌ large-scale experiments, from power‌ traces to secret key‌​‌ recovery (for most of​​ the instances) under a​​​‌ threshold of 18 hours,‌ targeting all three Kyber's‌​‌ security levels. Our average​​ rate of success across​​​‌ all security level is‌ more than 96%.

On‌​‌ the Success Rate of​​ Simple Side-Channel Attacks Against​​​‌ Masking with Unlimited Attack‌ Traces

Participants: Aymeric Hiltenbrand‌​‌, Julien Eynard,​​ Romain Poussier.

23​​​‌

In this work, we‌ investigate how the masking‌​‌ countermeasure affects the success​​ rate of simple attacks.​​​‌ To this end, we‌ provide theoretical, simulated, and‌​‌ practical experiments. Interestingly, we​​ will see that masking​​​‌ can allow us to‌ asymptotically recover more information‌​‌ on the secret than​​ in the case of​​​‌ an unprotected implementation, depending‌ on the masking type.‌​‌ We will see that​​ this is true for​​​‌ masking encodings that add‌ non-linearity with respect to‌​‌ the leakages, such as​​ arithmetic masking, while it​​​‌ is not for Boolean‌ masking. We believe this‌​‌ context provides interesting results,​​ as the average information​​​‌ of arithmetic encoding is‌ proven less informative than‌​‌ the Boolean one.

GnuZero:​​ A Compiler-Based Zeroization Static​​​‌ Detection Tool for the‌ Masses

Participants: Pierre-Alain Fouque‌​‌, Pierrick Philippe.​​

25

Coding standards for​​​‌ secure programming recommend "scrubbing"‌ sensitive data once it‌​‌ is no longer needed;​​ otherwise, secrets may be​​​‌ recovered, as illustrated in‌ the Heartbleed attack. Despite‌​‌ being an effective software-based​​ countermeasure, zeroization, i.e., overwriting​​​‌ with zeroes, turns out‌ to be challenging and‌​‌ error-prone. Current verification approaches​​ suffer from scalability or​​​‌ precision issues when applied‌ to production software in‌​‌ practice. In this paper,​​ we put forward the​​​‌ GCC Static Analyzer (GSA),‌ which is a symbolic‌​‌ execution engine for error​​ finding. Specifically, we extend​​​‌ the GSA to build‌ GnuZero; our automated tool‌​‌ that detects missing zeroization​​​‌ for all stack/heap variables​ storing sensitive data, either​‌ directly or by derivation.​​ Our experiments confirm GnuZero​​​‌ efficiency and effectiveness in​ verifying real-world benchmarks. In​‌ particular, GnuZero passes all​​ the relevant Juliet’s test​​​‌ programs, namely associated to​ the MITRE’s CWE-244 and​‌ CWE-226. In addition, GnuZero​​ succeeds in identifying new​​​‌ vulnerabilities in open-source cryptographic​ modules.

9 Bilateral contracts​‌ and grants with industry​​

9.1 Bilateral Grants with​​​‌ Industry

  • Resque: (T0:​ 09/2022 –> 08/2026)

    BPi​‌ France project.

    Led by​​ Thales.

    Participants: Pierre-Alain Fouque​​​‌, Guilhem Niot,​ Daniel De Almeida Braga​‌, Damien Marion,​​ Gaël Claudel.

    Participating​​​‌ entities on the industrial​ side: Thales SIX and​‌ DIS, TheGreenBow, CryptoExperts, CryptoNext.​​ Participating entities on the​​​‌ public side: Inria, ANSSI.​

    In this project, Inria​‌ is represented by two​​ teams: Capsule (Inria Rennes),​​​‌ with Pierre-Alain Fouque as​ the coordinator; and Cascade​‌ (Inria Paris), with Céline​​ Chevalier as collaborator.

    Resque​​​‌ project, "Résilience Quantique" aims​ at combining two use-cases​‌ allowing the construction of​​ two software and hardware​​​‌ components: i) VPN [virtual​ private network] hybrid and​‌ agile and a HSM​​ [hardware security module] robust​​​‌ and efficient, providing the​ security of exchanged information.​‌ The cryptographic agility will​​ allow to perform regular​​​‌ and continuous updates of​ the post-quantum algorithms.

  • Ascon-CAT​‌: (T0: 10/2024 –>​​ 09/2027)

    Participants: André Schrottenloher​​​‌, Aurel Pichollet–Mugnier.​

    AID “RAPID” project.

    Coordinated​‌ by Alice&Bob.

    Industrial partners:​​ Alice&Bob, Thales SIX. Academic​​​‌ partners: Inria.

    The goal​ of this project is​‌ to perform an integrated​​ quantum security analysis of​​​‌ the lightweight symmetric primitive​ ASCON, recently selected as​‌ a NIST standard. The​​ project will combine the​​​‌ development and analysis of​ new quantum algorithms, as​‌ well as a precise​​ estimation of the resources​​​‌ needed to run them,​ and a study of​‌ implementations in the “cat​​ qubits” platform which is​​​‌ developed by Alice&Bob.

10​ Partnerships and cooperations

10.1​‌ International initiatives

10.1.1 Visits​​ to international teams

Research​​​‌ stays abroad
Patrick Derbez​
  • Visited institution:
    Nanyang Technological​‌ University (NTU)
  • Country:
    Singapore​​
  • Dates:
    14/12/2025 – 17/12/2025​​​‌
  • Context of the visit:​
    Collaboration with Prof. Thomas​‌ Peyrin on two particular​​ topics: automated cryptanalysis and​​​‌ cryptanalysis of deep neural​ networks. I was also​‌ invited to give a​​ talk on a recent​​​‌ submitted paper about the​ cryptanalysis of deep neural​‌ networks with non-linear activation​​ functions.
  • Mobility program/type of​​​‌ mobility:
    research stay
André​ Schrottenloher
  • Visited institution:
    NTT​‌ Social Informatics Laboratories
  • Country:​​
    Japan
  • Dates:
    25 –​​​‌ 29 September 2025
  • Context​ of the visit:
    Visiting​‌ Akinori Hosoyamada, who is​​ a regular collaborator on​​​‌ the topic of quantum​ symmetric cryptanalysis.

10.2 National​‌ initiatives

  • The PQTLS (01/2022​​ –> 12/2027)

    Participants: Alexandre​​​‌ Wallet, Pierre-Alain Fouque​, André Schrottenloher,​‌ Yixin Shen, Clémence​​ Chevignard, Damien Marion​​​‌.

    Post-quantum padlock for​ web browser

    PEPR Quantique​‌

    Partners: GREYC (Caen), ENS​​ Lyon, Inria GRACE, Inria​​​‌ Cosmiq, Inria Prosecco, Inria​ Caramba, Inria Lfant, Inria​‌ Capsule, UVSQ, Cryptis, ARCAD,​​ SESAM, CEA LETI, University​​​‌ of Rouen, Rennes, Bordeaux.​

    The famous "padlock" appearing​‌ in browsers when one​​ visits websites whose address​​ is preceded by "https"​​​‌ relies on cryptographic primitives‌ that would not withstand‌​‌ a quantum computer. This​​ integrated project aims to​​​‌ develop in 5 years‌ post-quantum primitives in a‌​‌ prototype of "post-quantum lock"​​ that will be implemented​​​‌ in an open source‌ browser. The evolution of‌​‌ cryptographic standards has already​​ started, the choice of​​​‌ new primitives will be‌ made quickly, and the‌​‌ transition will be made​​ in the next few​​​‌ years. The objective is‌ to play a driving‌​‌ role in this evolution​​ and to make sure​​​‌ that the French actors‌ of post-quantum cryptography, already‌​‌ strongly involved, are able​​ to influence the cryptographic​​​‌ standards of the decades‌ to come.

  • Cryptanalyse (12/2023‌​‌ –> 12/2028)

    Participants: Patrick​​ Derbez, Aurore Guillevic​​​‌, André Schrottenloher.‌

    PEPR Cybersécurité

    Partners: Inria‌​‌ GRACE, Inria Cosmiq, Almasty,​​ Inria Caramba, Inria Lfant,​​​‌ Inria Capsule, Crypto, Eco,‌ Canari, UGA.

    The Cryptanalyse‌​‌ project focuses on the​​ study and standardization of​​​‌ cryptographic primitives. Modern cryptography‌ has become an indispensable‌​‌ tool for securing personal,​​ commercial and institutional communications.​​​‌ This project will provide‌ an estimate of the‌​‌ difficulties involved in solving​​ the underlying problems, and​​​‌ deduce the level of‌ security conferred by the‌​‌ use of these primitives.​​ The aim is to​​​‌ evaluate the security of‌ cryptographic algorithms.

  • CROWD (2023‌​‌ –> 2027).

    Participants: Pierre-Alain​​ Fouque, André Schrottenloher​​​‌, Clémence Chevignard.‌

    Code-based practical cryptography

    ANR-DFG‌​‌

    Partners: TU Munich, IRMAR​​ (Rennes), Inria (Rennes)

    The​​​‌ aim of this project‌ is the examination of‌​‌ skew metrics and their​​ application in cryptography. These​​​‌ metrics can be considered‌ as a generalization of‌​‌ the so-called rank metric,​​ which has significant applications​​​‌ in coding theory, cryptography,‌ data storage, and network‌​‌ coding. The connection of​​ these metrics lies in​​​‌ the non-commutativity of Euclidean‌ rings, called Ore rings,‌​‌ which extend the classical​​ notation of commutative polynomial​​​‌ rings by 'skewing' (twisting)‌ multiplication. These operations allow‌​‌ the development of metrics​​ and new codes with​​​‌ efficient arithmetic operations. This‌ holds promise for secure‌​‌ and efficient cryptographic implementations.​​ Three avenues are explored:​​​‌ 1) investigates the foundations‌ of algebraic codes in‌​‌ these skew-metrics; 2) design​​ novel decoding algorithms and​​​‌ cryptographic schemes from these‌ codes, and assess their‌​‌ security from a cryptanalytic​​ and side-channel point of​​​‌ view; 3) produce practically‌ efficient implementation of core‌​‌ cryptographic primitive, such as​​ digital signatures.

  • ANR IDROMEL​​​‌ (2021 –> 2025)

    Participants:‌ Damien Marion.

    Improving‌​‌ the Design of secure​​ systems by a Reduction​​​‌ Of Micro-architectural Effects on‌ side-channeL Attacks

    Partners: LAAS-CNRS,‌​‌ LIP6, CEA, ARM, IRISA​​

    The IDROMEL project aims​​​‌ to contribute to the‌ design of secure systems‌​‌ against side-channel attacks based​​ on power and electromagnetic​​​‌ observations, for a wide‌ range of computing systems‌​‌ (from IoT devices to​​ mobile phones). IDROMEL will​​​‌ investigate the impact of‌ the processor micro-architecture on‌​‌ power and electromagnetic side-channel​​ attacks as a key​​​‌ concern for the design‌ of secure systems.

  • ANR‌​‌ OREO (2023 –> 2026)​​

    Participants: Patrick Derbez,​​​‌ Andre Schrottenloher.

    MILP‌ for Cryptography

    Partners: Univ‌​‌ Rennes, UVSQ, Loria

    In​​​‌ symmetric-key cryptography, a popular​ technique for proving resistance​‌ against classical attacks is​​ to model the behaviour​​​‌ of the cipher as​ a Mixed Integer Linear​‌ Programming (MILP) problem and​​ solve it by some​​​‌ MILP solver. This method​ was applied for the​‌ first time by Mouha​​ et al. [MWGP11] and​​​‌ by Wu and Wang​ [WW11] for finding the​‌ minimum number of differentially​​ and linearly active Sboxes​​​‌ and provides in such​ a way a proof​‌ of resistance against these​​ two classical attacks. Since​​​‌ then, the use of​ MILP not only by​‌ designers but also by​​ cryptanalysts has increased, the​​​‌ advantage being that many​ cryptanalytic problems are relatively​‌ easy to translate into​​ linear constraints (typically on​​​‌ bits) and available solvers​ (e.g. Gurobi, CPLEX) are​‌ most often very efficient​​ to solve them.

    Currently,​​​‌ MILP solvers are mainly​ used for differential cryptanalysis,​‌ including the search for​​ sophisticated boomerang distinguishers, and​​​‌ for integral cryptanalysis by​ exhausting division trails on​‌ a cipher. But we​​ are reaching a point​​​‌ where describing the problem​ into a MILP model​‌ and solving it naively​​ is not enough. Thus​​​‌ there are many open​ problems related to MILP​‌ applied to cryptography and​​ the aim of this​​​‌ new ANR project is​ to tackle them. Our​‌ main objective is to​​ handle more complex cryptographic​​​‌ problems, relying on both​ a theoretical work on​‌ cryptanalysis techniques and an​​ improvement of MILP models.​​​‌ The project is composed​ of 4 axis: handling​‌ more complex cryptographic problems​​ using MILP solvers, automatically​​​‌ searching for key-recovery attacks,​ side-channels cryptanalysis and conception​‌ of cryptographic primitives.

  • ANR​​ JCJC QATS (2025 –>​​​‌ 2029): Quantum Attacks and​ new Tools for Symmetric​‌ Cryptanalysis

    Participants: Alisée Lafontaine​​, Andre Schrottenloher.​​​‌

    Nowadays, symmetric cryptanalysis relies​ heavily on automatic tools.​‌ These tools model the​​ search for an attack​​​‌ as an optimization problem,​ which is solved using​‌ off-the-shelf solvers. Regarding quantum​​ security, at the moment,​​​‌ only a few quantum​ attacks have been integrated​‌ into such tools. Besides,​​ significant human effort is​​​‌ still required to determine​ precisely the complexity of​‌ the attack, especially in​​ the quantum setting.

    The​​​‌ goal of the ANR​ JCJC QATS project is​‌ to synthesize a single​​ toolchain to output fully​​​‌ specified quantum attack algorithms,​ and their complexities. Primitives​‌ such as block ciphers​​ and hash functions will​​​‌ be analyzed, starting from​ well-established designs and moving​‌ towards more recent ones.​​ This toolchain is expected​​​‌ to simplify the study​ of quantum attacks, especially​‌ the computation of their​​ complexity. We aim to​​​‌ produce a toolbox where​ the quantum security of​‌ a primitive can be​​ estimated with only basic​​​‌ knowledge of symmetric cryptanalysis​ and quantum algorithms. This​‌ would be helpful for​​ designers of new algorithms,​​​‌ and more generally, cryptographers​ interested in quantum security​‌ estimates.

11 Dissemination

11.1​​ Promoting scientific activities

11.1.1​​​‌ Scientific events: organisation

Member of the​​​‌ organizing committees

Participants: Aurore‌ Guillevic, Damien Marion‌​‌, André Schrottenloher,​​ Yixin Shen.

  • Séminaire​​​‌ CRYPTO (IRMAR, IRISA, Rennes):‌ Aurore Guillevic and Damien‌​‌ Marion until September 2025,​​ André Schrottenloher, Yixin Shen.​​​‌

11.1.2 Scientific events: selection‌

Chair of conference program‌​‌ committees

Participants: Pierre-Alain Fouque​​, André Schrottenloher.​​​‌

Member‌​‌ of the conference program​​ committees

Participants: Daniel de​​​‌ Almeida Braga, Patrick‌ Derbez, Aurore Guillevic‌​‌, André Schrottenloher,​​ Yixin Shen.

Reviewer

The team members​​ regularly serve as sub-reviewers​​​‌ for the IACR conferences.‌ For anonymity reasons, the‌​‌ details are not provided.​​

11.1.3 Journal

Member of​​​‌ the editorial boards

Participants:‌ Patrick Derbez, André‌​‌ Schrottenloher.

Reviewer -‌​‌ reviewing activities

Team members​​ regularly review papers submitted​​​‌ to international journals such‌ as Designs, Codes, and‌​‌ Cryptography (DCC), Finite Fields​​ and their Applications (FFA),​​​‌ Journal of Cryptology. For‌ anonymity reasons, the details‌​‌ are not provided.

11.1.4​​ Invited talks

Participants: Clémence​​​‌ Chevignard, Pierrick Dartois‌, Patrick Derbez,‌​‌ Pierre-Alain Fouque, André​​ Schrottenloher, Yixin Shen​​​‌.

11.1.5 Scientific expertise

Participants:​​ Pierre-Alain Fouque, Aurore​​​‌ Guillevic, André Schrottenloher​, Yixin Shen.​‌

  • Pierre-Alain Fouque was the​​ president of the selection​​​‌ committee for a professor​ position in section 27​‌ at the Université de​​ Rennes.
  • Aurore Guillevic was​​​‌ a member of the​ selection committee for an​‌ assistant professor position (MCF252186)​​ in section 27 at​​​‌ Université de Versailles Saint​ Quentin (UVSQ).
  • Yixin Shen​‌ was a jury member​​ of the selection committee​​​‌ for INRIA CRCN/ISFP researcher​ position at the INRIA​‌ Saclay center.
  • André Schrottenloher​​ reviewed projects for ID4Mobility​​​‌ and for the Czech​ Science Foundation.

11.1.6 Research​‌ administration

Participants: Aurore Guillevic​​, Yixin Shen.​​​‌

  • Aurore Guillevic is in​ charge of the young​‌ researchers (mission jeunes chercheurs)​​ and at the Commission​​​‌ Personnel.
  • Yixin Shen is​ at the Commission Délégation​‌ Inria.

11.2 Teaching -​​ Supervision - Juries -​​​‌ Educational and pedagogical outreach​

11.2.1 Teaching

Participants: Daniel​‌ de Almeida Braga,​​ Patrick Derbez, Pierre-Alain​​​‌ Fouque, Aurore Guillevic​, Damien Marion,​‌ André Schrottenloher.

  • Master:​​ André Schrottenloher, Enjeux de​​​‌ la cryptographie post-quantique, 3​ hours conference, Centrale- Supélec​‌ Rennes.
  • Master: André Schrottenloher,​​ Cryptanalyse , 7.5 hours​​​‌ lectures, 7.5 hours lab​ sessions, M2, University of​‌ Rennes.
  • Bachelor: André Schrottenloher,​​ Introduction à la Cryptologie,​​​‌ 9 hours lectures, ENS​ de Rennes.
  • Master: Aurore​‌ Guillevic and Gaël Claudel,​​ Advanced Course in Cryptography​​​‌ for security (BCS), 16.5​ hours lab sessions, M2,​‌ University of Rennes, France;​​
  • Master: Aurore Guillevic, Mathematics​​​‌ for security (MSEC), 12​ hours lectures, 2 ×​‌ 12 hours lab sessions,​​ M1, University of Rennes,​​​‌ France;
  • Master: Aurore Guillevic,​ Unix refresher crash course,​‌ 3 hours lab sessions,​​ M1, University of Rennes,​​​‌ France;
  • Master: Pierre-Alain Fouque,​ Basics for Cryptography, 22h​‌ lectures, M1, University of​​ Rennes, France;
  • Master: Pierre-Alain​​​‌ Fouque, Advanced Cryptography, 16h​ lectures, M2, University of​‌ Rennes, France;
  • Master: Pierre-Alain​​ Fouque, Security Proof, 12h​​​‌ lectures, M2, University of​ Rennes, France;
  • Bachelor: Daniel​‌ De Almeida Braga, Introduction​​ à la Securité (ISE),​​​‌ 1.5 hour lecture, 3​ hours lab session, L1,​‌ University of Rennes, France;​​
  • Bachelor: Daniel De Almeida​​​‌ Braga, Enjeux de Sécurité​ (ESEC), 7.5 hours lectures,​‌ L3, University of Rennes,​​ France;
  • Master: Daniel De​​​‌ Almeida Braga, Security Project,​ 24h project supervision, M1,​‌ University of Rennes, France;​​
  • Master: Daniel De Almeida​​​‌ Braga, Low Level Programming,​ 19.5 hours lab sessions,​‌ M1, University of Rennes,​​ France;
  • Master: Daniel De​​​‌ Almeida Braga, Cybersécurité: Menaces​ et organisations, hygiène numérique​‌ (SENV), 6 hours lectures,​​ 10.5 hours lab sessions,​​​‌ M1, University of Rennes,​ France;
  • Master: Daniel De​‌ Almeida Braga, Sécurité avancée​​ des SI d'entreprise (SSYS2),​​​‌ 18 hours lectures, 36​ hours lab sessions, M2,​‌ University of Rennes, France;​​
  • Master: Daniel De Almeida​​​‌ Braga, Introduction au pentest,​ 6 hours lectures, 12​‌ hours lab sessions, M1,​​ University of Rennes, France;​​
  • Daniel De Almeida Braga:​​​‌ Co-director of the Master‌ 2 CISO (RSSI), University‌​‌ of Rennes, France.
  • Bachelor:​​ Damien Marion, numérique éco-responsable​​​‌ (NEC2), 7.5 hours lectures,‌ 9 hours lab sessions,‌​‌ 9 hours seminars, 9​​ hours projects, L2, University​​​‌ of Rennes, France;
  • Bachelor:‌ Damien Marion, enjeux sociétaux‌​‌ et empreinte écologique du​​ numérique (3EN), 6 hours​​​‌ lectures, 12 hours projects,‌ L3, University of Rennes,‌​‌ France;
  • Master: Damien Marion,​​ databases security (SBD), 33​​​‌ hours lab sessions, M1,‌ University of Rennes, France;‌​‌
  • Master: Damien Marion, cryptography​​ and privacy (SDATA), 12​​​‌ hours lectures, 21 hours‌ lab sessions, 10.5 hours‌​‌ seminars, M1, University of​​ Rennes, France;
  • Master: Damien​​​‌ Marion, secured implementations for‌ cryptography (SIMP), 3.5 hours‌​‌ lectures, 39 hours lab​​ sessions, M2, University of​​​‌ Rennes, France;
  • Master: Damien‌ Marion, research project, 24‌​‌ hours project supervision, M1,​​ University of Rennes, France;​​​‌
  • Damien Marion: contact regarding‌ teaching activities of the‌​‌ ecological transition in computer​​ science at the University​​​‌ of Rennes, France;
  • Master:‌ Patrick Derbez, Cryptanalyse, 7.5‌​‌ hours lectures, 7.5 hours​​ lab sessions, M2, University​​​‌ of Rennes.
  • Bachelor: Patrick‌ Derbez, Introduction à la‌​‌ programmation, 15 hours lectures,​​ 16.5 hours lab sessions,​​​‌ L1, University of Rennes.‌
  • Bachelor: Patrick Derbez, Introduction‌​‌ à la Cryptologie, 3​​ hours lectures, 3 hours​​​‌ lab sessions, L1, University‌ of Rennes.

11.2.2 Supervision‌​‌

Participants: Daniel de Almeida​​ Braga, Patrick Derbez​​​‌, Pierre-Alain Fouque,‌ Aurore Guillevic, Damien‌​‌ Marion, André Schrottenloher​​, Yixin Shen.​​​‌

  • PhD: Phuong Hoa Nguyen,‌ MILP and symmetric-key cryptanalysis‌​‌33, started October​​ 2021, defended February 2025.​​​‌ Supervisors: Patrick Derbez and‌ Pierre-Alain Fouque.
  • PhD: Paul‌​‌ Kirchner, Cryptanalysis of public-key​​ cryptography32, defended​​​‌ May 2025. Supervisors: Pierre-Alain‌ Fouque, with Aurore Guillevic‌​‌ in 2024–2025.
  • PhD: Pierrick​​ Philippe, Secrets in Compiler:​​​‌ Detection of Secret-related Weaknesses‌ in GCC Static Analyzer‌​‌, started October 2022,​​ defended December 2025. Supervisors:​​​‌ Mohamed Sabt (IRISA) and‌ Pierre-Alain Fouque.
  • PhD in‌​‌ progress: Clémence Chevignard, Module-LIP:​​ réductions, cryptanalyse, algorithmes,​​​‌ started November 2023. Supervisors:‌ Pierre-Alain Fouque, Alexandre Wallet‌​‌ and Rémi Giraud (Qualcomm).​​
  • PhD in progress: Mathieu​​​‌ Degré, Nouveaux modèles MILP‌ adaptés aux problèmes cryptographiques‌​‌, started January 2024.​​ Supervisors: Patrick Derbez, André​​​‌ Schrottenloher.
  • PhD in progress:‌ Aurel Pichollet–Mugnier, Security of‌​‌ ASCON and Lightweight Symmetric​​ Primitives against Quantum Attackers​​​‌, started November 2024.‌ Supervisors: Patrick Derbez, André‌​‌ Schrottenloher, Zoé Amblard (Thales​​ SIX)
  • PhD in progress:​​​‌ Baptiste Germon, Independence hypothesis‌ in differential cryptanalysis,‌​‌ started October 2024. Supervisors:​​ Patrick Derbez, Christina Boura​​​‌ (IRIF)
  • PhD in progress:‌ Roderick Asselineau, Cryptanalyse d'algorithmes‌​‌ symétriques utilisés dans la​​ vie réelle, started​​​‌ April 2025. Supervisors: Patrick‌ Derbez, Pierre-Alain Fouque, Brice‌​‌ Minaud (Inria Paris).
  • PhD​​ in progress: Bastien Michel,​​​‌ Optimisation de la cryptanalyse‌ de primitives symétrique,‌​‌ started October 2024. Supervisors:​​ Patrick Derbez, Maria Naya-Plasencia​​​‌ (Inria Paris).
  • PhD in‌ progress: Marie Euler, Outils‌​‌ pour la cryptanalyse et​​ la conception de primitives​​​‌ cryptographiques, VAE. Supervisors:‌ Patrick Derbez.
  • PhD in‌​‌ progress: Gaël Claudel, Analyse​​ des attaques par canaux​​​‌ auxiliaires de schémas de‌ signature post quantique :‌​‌ approches combinées. Supervisors:​​​‌ Patrick Derbez, Damien Marion,​ Aurore Guillevic, Benoît Gérard​‌ (ANSSI).
  • PhD in progress:​​ Aymeric Hiltenbrand, Attaques par​​​‌ canaux auxiliaires sur la​ cryptographie post- quantique,​‌ from December 2023. Supervisors:​​ Guenael Renault (ANSSI), Pierre-Alain​​​‌ Fouque, Romain Poussier (ANSSI),​ Damien Marion.
  • PhD in​‌ progress: Guilhem Niot, Threshold​​ Post-Quantum Cryptography. Supervisors:​​​‌ Pierre- Alain Fouque and​ Thomas Prest (PQShield).
  • PhD​‌ in progress: Mathias Boucher.​​ Improved Quantum Cryptanalysis on​​​‌ Lattices, started September​ 2025. Supervisors: Yixin Shen​‌ and Pierre-Alain Fouque.
  • PhD​​ in progress: Alisée Lafontaine,​​​‌ New tools for quantum​ symmetric cryptanalysis, started​‌ October 2025. Supervisors: André​​ Schrottenloher and Patrick Derbez.​​​‌
  • PhD in progress: Paul​ Delhom, Signatures Avancées Post-Quantiques​‌, started November 2025.​​ Supervisors: Corentin Jeudy (Orange),​​​‌ Olivier Sanders (Orange) and​ Pierre-Alain Fouque.
  • PhD in​‌ progress: Théo Goureau, ARMADA:​​ ARM Microarchitectural Attacks Discovery​​​‌ and Analysis, started​ October 2025. Supervisors: Daniel​‌ De Almeida Braga, Pierre-Alain​​ Fouque, Guillaume Hiet (SUSHI)​​​‌ and Thomas Rokicki (SUSHI).​
  • Internship: Mathias Boucher (M2)​‌ Regev’s reduction on a​​ family of easily decodable​​​‌ lattices (March-July 2025). Supervisor:​ Yixin Shen.
  • Internship: Alisée​‌ Lafontaine (M2) New tools​​ for quantum symmetric cryptanalysis​​​‌ (March-September 2025). Supervisor: André​ Schrottenloher.
  • Internship: Babacar Ndiaye​‌ (M2 Université de Limoges)​​ Polynomial selection for the​​​‌ Tower Number Field Sieve​ (April-August 2025). Supervisor: Aurore​‌ Guillevic.
  • Internship: Hubert de​​ Groote (M2 MPRI) Security​​​‌ proof for Falcon sampler​ (April-August 2025). Supervisor: Pierre-Alain​‌ Fouque.
  • Internship: Adrian Lagasse​​ (M1 Université de Rennes)​​​‌ Analysis and Exploitation of​ Intel's Instruction Prefetcher (May-August​‌ 2025). Supervisor: Daniel De​​ Almeida Braga.

11.2.3 Juries​​​‌

Participants: Pierre-Alain Fouque,​ Aurore Guillevic, André​‌ Schrottenloher, Yixin Shen​​.

  • Pierre-Alain Fouque was​​​‌ a reviewer of the​ PhD thesis of Philipp​‌ Gajland (June 4, 2025,​​ Bochum-Universität, Germany).
  • Pierre-Alain Fouque​​​‌ was a reviewer of​ the HDR thesis of​‌ Yann Rotella (February 6,​​ 2025, Université de Versailles-Saint-Quentin-en-Yvelines,​​​‌ France).
  • Pierre-Alain Fouque was​ the President of the​‌ PhD thesis of Tristan​​ Claverie (June 6, 2025,​​​‌ Université de Rennes, France).​
  • Pierre-Alain Fouque was an​‌ examiner of the PhD​​ thesis of Paul Kirchner​​​‌ (May 22, 2025, Université​ de Rennes, France).
  • Pierre-Alain​‌ Fouque was the President​​ of the PhD thesis​​​‌ of Charles Meyer-Hilfiger (September​ 30, 2025, Université de​‌ Sorbonne University, France).
  • Pierre-Alain​​ Fouque was a reviewer​​​‌ of the PhD thesis​ of Nicolas Bon (November​‌ 14, 2025, École normale​​ supérieure, France).
  • Pierre-Alain Fouque​​​‌ was a reviewer of​ the PhD thesis of​‌ Henry Bambury (November 18,​​ 2025, École normale supérieure,​​​‌ France).
  • Pierre-Alain Fouque was​ an examiner of the​‌ PhD thesis of Guirec​​ Lebrun (December 1, 2025,​​​‌ École normale supérieure, France).​
  • Pierre-Alain Fouque was the​‌ President of the PhD​​ thesis of Pierre Pébereau​​​‌ (December 16, 2025, Université​ de Sorbonne University, France).​‌
  • Pierre-Alain Fouque was an​​ examiner of the PhD​​​‌ thesis of Pierrick Philippe​ (December 10, 2025, Université​‌ de Rennes, France).
  • Pierre-Alain​​ Fouque was a reviewer​​​‌ of the PhD thesis​ of Viet-Sang Nguyen (December​‌ 19, 2025, Université de​​ Saint-Étienne, France).
  • André Schrottenloher​​​‌ was a reviewer of​ the PhD thesis of​‌ Nathalie Lang (July 14,​​ 2025, Bauhaus-Universität Weimar, Germany).​​
  • Aurore Guillevic was a​​​‌ reviewer of the PhD‌ thesis of François Palma‌​‌ (December 12, 2025, Université​​ de Toulon, France).
  • Yixin​​​‌ Shen was an examiner‌ of the PhD thesis‌​‌ of Joseph Cunningham (September​​ 5, 2025, Université Libre​​​‌ de Bruxelles, Belgique).

11.2.4‌ Educational and pedagogical outreach‌​‌

Participants: Daniel de Almeida​​ Braga, Patrick Derbez​​​‌, Pierre-Alain Fouque,‌ Yixin Shen.

  • Yixin‌​‌ Shen gave a talk​​ at the ENS Rennes​​​‌ student research seminar about‌ "Collision finding, random walks‌​‌ and quantum algorithms" in​​ November 2025.
  • Daniel De​​​‌ Almeida Braga organized a‌ Capture The Flag (CTF)‌​‌ event for students at​​ University of Rennes in​​​‌ April 2025, to bridge‌ bachelor and master, with‌​‌ around 70 participants.
  • Patrick​​ Derbez gave a 3​​​‌ hours lecture on "Differential‌ Cryptanalysis" at the summer‌​‌ school colocated to the​​ SAC 2025 conference in​​​‌ August 2025. SAC 2025‌ Summer School
  • Pierre-Alain Fouque‌​‌ gave a 3 hours​​ lecture on "Quantum algorithms​​​‌ for cryptanalysis" at the‌ CEMRACS 2025 in July‌​‌ 2025. CIRM Luminy

11.3​​ Popularization

11.3.1 Productions (articles,​​​‌ videos, podcasts, serious games,‌ ...)

Participants: Damien Marion‌​‌, Yixin Shen.​​

11.3.2 Participation in​​​‌ Live events

Participants: André‌ Schrottenloher, Yixin Shen‌​‌.

  • Yixin Shen and​​ André Schrottenloher were panelists​​​‌ of Amphi Métier R&D,‌ at Ecole polytechnique, Palaiseau,‌​‌ France, 2025

12 Scientific​​ production

12.1 Major publications​​​‌

12.2 Publications of the​​ year

International journals

International peer-reviewed​​ conferences

Conferences without proceedings​‌

Doctoral dissertations​​ and habilitation theses

Reports & preprints

12.3​​​‌ Cited publications

  • 37 inproceedings‌L.Léo Ducas and‌​‌ L. N.Ludo N.​​ Pulles. Does the​​​‌ Dual-Sieve Attack on Learning‌ with Errors even Work?‌​‌crypto # ~202314083​​lncsSanta Barbara, CA,​​​‌ USASpringerAugust 2023‌, 37--69URL: https://doi.org/10.1007/978-3-031-38548-3_2‌​‌DOIback to text​​
  • 38 inproceedingsD.Divesh​​​‌ Aggarwal, D.Daniel‌ Dadush, O.Oded‌​‌ Regev and N.Noah​​ Stephens-Davidowitz. Solving the​​​‌ Shortest Vector Problem in‌ 2 n Time Using‌​‌ Discrete Gaussian Sampling: Extended​​ Abstract.Proceedings of​​​‌ the Forty-Seventh Annual ACM‌ on Symposium on Theory‌​‌ of Computing, STOC 2015,​​ Portland, OR, USA, June​​​‌ 14-17, 20152015,‌ 733--742URL: https://doi.org/10.1145/2746539.2746606DOI‌​‌back to text
  • 39​​ inproceedingsG.Gorjan Alagic​​​‌, C.Chen Bai‌, J.Jonathan Katz‌​‌ and C.Christian Majenz​​. Post-Quantum Security of​​​‌ the Even-Mansour Cipher.‌EUROCRYPT (3)13277Lecture‌​‌ Notes in Computer Science​​Springer2022, 458--487​​​‌back to text
  • 40‌ inproceedingsZ.Zhenzhen Bao‌​‌, J.Jian Guo​​, D.Danping Shi​​​‌ and Y.Yi Tu‌. Superposition Meet-in-the-Middle Attacks:‌​‌ Updates on Fundamental Security​​ of AES-like Hashing.​​​‌CRYPTO (1)13507Lecture‌ Notes in Computer Science‌​‌Springer2022, 64--93​​back to text
  • 41​​​‌ inproceedingsG.Gilles Barthe‌, S.Sonia Belaïd‌​‌, G.Gaëtan Cassiers​​, P.-A.Pierre-Alain Fouque​​​‌, B.Benjamin Grégoire‌ and F.-X.François-Xavier Standaert‌​‌. maskVerif: Automated Verification​​ of Higher-Order Masking in​​​‌ Presence of Physical Defaults‌.ESORICS (1)11735‌​‌Lecture Notes in Computer​​ ScienceSpringer2019,​​​‌ 300--318back to text‌
  • 42 inproceedingsG.Gilles‌​‌ Barthe, S.Sonia​​​‌ Belaïd, F.François​ Dupressoir, P.-A.Pierre-Alain​‌ Fouque, B.Benjamin​​ Grégoire and P.-Y.Pierre-Yves​​​‌ Strub. Verified Proofs​ of Higher-Order Masking.​‌EUROCRYPT (1)9056Lecture​​ Notes in Computer Science​​​‌Springer2015, 457--485​back to text
  • 43​‌ inproceedingsG.Gilles Barthe​​, S.Sonia Belaïd​​​‌, F.François Dupressoir​, P.-A.Pierre-Alain Fouque​‌, B.Benjamin Grégoire​​, P.-Y.Pierre-Yves Strub​​​‌ and R.Rébecca Zucchini​. Strong Non-Interference and​‌ Type-Directed Higher-Order Masking.​​CCSACM2016,​​​‌ 116--129back to text​
  • 44 inproceedingsG.Gilles​‌ Barthe, S.Sonia​​ Belaïd, T.Thomas​​​‌ Espitau, P.-A.Pierre-Alain​ Fouque, B.Benjamin​‌ Grégoire, M.Mélissa​​ Rossi and M.Mehdi​​​‌ Tibouchi. Masking the​ GLP Lattice-Based Signature Scheme​‌ at Any Order.​​EUROCRYPT (2)10821Lecture​​​‌ Notes in Computer Science​Springer2018, 354--384​‌back to textback​​ to text
  • 45 inproceedings​​​‌G.Gilles Barthe,​ S.Sonia Belaïd,​‌ T.Thomas Espitau,​​ P.-A.Pierre-Alain Fouque,​​​‌ M.Mélissa Rossi and​ M.Mehdi Tibouchi.​‌ GALACTICS: Gaussian Sampling for​​ Lattice-Based Constant- Time Implementation​​​‌ of Cryptographic Signatures, Revisited​.CCSACM2019​‌, 2147--2164back to​​ text
  • 46 inproceedingsO.​​​‌Olivier Bernard and A.​Adeline Roux-Langlois. Twisted-PHS:​‌ Using the Product Formula​​ to Solve Approx-SVP in​​​‌ Ideal Lattices.ASIACRYPT​ (2)12492Lecture Notes​‌ in Computer ScienceSpringer​​2020, 349--380back​​​‌ to text
  • 47 article​G.Guido Bertoni,​‌ J.Joan Daemen,​​ S.Seth Hoffert,​​​‌ M.Michaël Peeters,​ G. V.Gilles Van​‌ Assche and R. V.​​Ronny Van Keer.​​​‌ Farfalle: parallel permutation-based cryptography​.IACR Trans. Symmetric​‌ Cryptol.201742017​​, 1--38back to​​​‌ text
  • 48 inproceedingsJ.-F.​Jean-François Biasse, T.​‌Thomas Espitau, P.-A.​​Pierre-Alain Fouque, A.​​​‌Alexandre Gélin and P.​Paul Kirchner. Computing​‌ Generator in Cyclotomic Integer​​ Rings - A Subfield​​​‌ Algorithm for the Principal​ Ideal Problem in L​‌ (1/2​​) and Application to​​​‌ the Cryptanalysis of a​ FHE Scheme.EUROCRYPT​‌ (1)10210Lecture Notes​​ in Computer Science2017​​​‌, 60--88back to​ text
  • 49 inproceedingsX.​‌Xavier Bonnetain, A.​​André Chailloux, A.​​​‌André Schrottenloher and Y.​Yixin Shen. Finding​‌ Many Collisions via Reusable​​ Quantum Walks - Application​​​‌ to Lattice Sieving.​EUROCRYPT (5)14008Lecture​‌ Notes in Computer Science​​Springer2023, 221--251​​​‌back to textback​ to text
  • 50 article​‌X.Xavier Bonnetain,​​ M.María Naya-Plasencia and​​​‌ A.André Schrottenloher.​ Quantum Security Analysis of​‌ AES.IACR Trans.​​ Symmetric Cryptol.20192​​​‌2019, 55--93URL:​ https://doi.org/10.13154/tosc.v2019.i2.55-93DOIback to​‌ text
  • 51 inproceedingsX.​​Xavier Bonnetain, A.​​​‌André Schrottenloher and F.​Ferdinand Sibleyras. Beyond​‌ Quadratic Speedups in Quantum​​ Attacks on Symmetric Schemes​​​‌.EUROCRYPT (3)13277​Lecture Notes in Computer​‌ ScienceSpringer2022,​​ 315--344back to text​​​‌
  • 52 inproceedingsK.Katharina​ Boudgoust, C.Corentin​‌ Jeudy, A.Adeline​​ Roux-Langlois and W.Weiqiang​​ Wen. Towards Classical​​​‌ Hardness of Module-LWE: The‌ Linear Rank Case.‌​‌ASIACRYPT (2)12492Lecture​​ Notes in Computer Science​​​‌Springer2020, 289--317‌back to text
  • 53‌​‌ inproceedingsC.Charles Bouillaguet​​, C.Claire Delaplace​​​‌ and M.Monika Trimoska‌. A Simple Deterministic‌​‌ Algorithm for Systems of​​ Quadratic Polynomials over 2​​​‌ .SOSASIAM2022‌, 285--296back to‌​‌ text
  • 54 inproceedingsC.​​Charles Bouillaguet, P.​​​‌Patrick Derbez and P.-A.‌Pierre-Alain Fouque. Automatic‌​‌ Search of Attacks on​​ Round-Reduced AES and Applications​​​‌.CRYPTO6841Lecture‌ Notes in Computer Science‌​‌Springer2011, 169--187​​back to text
  • 55​​​‌ inproceedingsD. D.Daniel‌ De Almeida Braga,‌​‌ P.-A.Pierre-Alain Fouque and​​ M.Mohamed Sabt.​​​‌ Dragonblood is Still Leaking:‌ Practical Cache-based Side-Channel in‌​‌ the Wild.ACSAC​​ACM2020, 291--303​​​‌back to text
  • 56‌ inproceedingsD. D.Daniel‌​‌ De Almeida Braga,​​ N.Natalia Kulatova,​​​‌ M.Mohamed Sabt,‌ P.-A.Pierre-Alain Fouque and‌​‌ K.Karthikeyan Bhargavan.​​ From Dragondoom to Dragonstar:​​​‌ Side-channel Attacks and Formally‌ Verified Implementation of WPA3‌​‌ Dragonfly Handshake.EuroS&P​​IEEE2023, 707--723​​​‌back to text
  • 57‌ inproceedingsZ.Zvika Brakerski‌​‌, A.Adeline Langlois​​, C.Chris Peikert​​​‌, O.Oded Regev‌ and D.Damien Stehlé‌​‌. Classical hardness of​​ learning with errors.​​​‌STOCACM2013,‌ 575--584back to text‌​‌
  • 58 articleY.Yanlin​​ Chen, K.-M.Kai-Min​​​‌ Chung and C.-Y.Ching-Yi‌ Lai. Space-efficient classical‌​‌ and quantum algorithms for​​ the shortest vector problem​​​‌.Quantum Information &‌ Computation183&42018‌​‌, 285--306URL: http://www.rintonpress.com/xxqic18/qic-18-34/0285-0306.pdf​​back to text
  • 59​​​‌ inproceedingsJ. H.Jung‌ Hee Cheon, P.-A.‌​‌Pierre-Alain Fouque, C.​​Changmin Lee, B.​​​‌Brice Minaud and H.‌Hansol Ryu. Cryptanalysis‌​‌ of the New CLT​​ Multilinear Map over the​​​‌ Integers.EUROCRYPT (1)‌9665Lecture Notes in‌​‌ Computer ScienceSpringer2016​​, 509--536back to​​​‌ text
  • 60 inproceedingsC.‌Chitchanok Chuengsatiansup, T.‌​‌Thomas Prest, D.​​Damien Stehlé, A.​​​‌Alexandre Wallet and K.‌Keita Xagawa. ModFalcon:‌​‌ Compact Signatures Based On​​ Module-NTRU Lattices.AsiaCCS​​​‌ACM2020, 853--866‌back to text
  • 61‌​‌ inproceedingsD.Daniel Collins​​, L.Loïs Huguenin-Dumittan​​​‌, N. K.Ngoc‌ Khanh Nguyen, N.‌​‌Nicolas Rolin and S.​​Serge Vaudenay. K-waay:​​​‌ fast and deniable post-quantum‌ X3DH without ring signatures‌​‌.Proceedings of the​​ 33rd USENIX Conference on​​​‌ Security SymposiumSEC '24‌USAPhiladelphia, PA, USA‌​‌USENIX Association2024back​​ to text
  • 62 inproceedings​​​‌P.Patrick Derbez and‌ P.-A.Pierre-Alain Fouque.‌​‌ Automatic Search of Meet-in-the-Middle​​ and Impossible Differential Attacks​​​‌.Advances in Cryptology‌ - CRYPTO 2016 -‌​‌ 36th Annual International Cryptology​​ Conference, Santa Barbara, CA,​​​‌ USA, August 14-18, 2016,‌ Proceedings, Part II9815‌​‌Lecture Notes in Computer​​ ScienceSpringer2016,​​​‌ 157--184DOIback to‌ text
  • 63 inproceedingsP.‌​‌Patrick Derbez and P.-A.​​Pierre-Alain Fouque. Exhausting​​​‌ Demirci-Selçuk Meet-in-the-Middle Attacks Against‌ Reduced-Round AES.FSE‌​‌8424Lecture Notes in​​​‌ Computer ScienceSpringer2013​, 541--560back to​‌ text
  • 64 articleC.​​Christoph Dobraunig, M.​​​‌Maria Eichlseder, F.​Florian Mendel and M.​‌Martin Schläffer. Ascon​​ v1.2: Lightweight Authenticated Encryption​​​‌ and Hashing.J.​ Cryptol.3432021​‌, 33back to​​ text
  • 65 inproceedingsV.​​​‌Vivien Dubois, P.-A.​Pierre-Alain Fouque, A.​‌Adi Shamir and J.​​Jacques Stern. Practical​​​‌ Cryptanalysis of SFLASH.​CRYPTO4622Lecture Notes​‌ in Computer ScienceSpringer​​2007, 1--12back​​​‌ to text
  • 66 inproceedings​A.Alexandre Duc,​‌ S.Stefan Dziembowski and​​ S.Sebastian Faust.​​​‌ Unifying Leakage Models: From​ Probing Attacks to Noisy​‌ Leakage.EUROCRYPT8441​​Lecture Notes in Computer​​​‌ ScienceSpringer2014,​ 423--440back to text​‌
  • 67 inproceedingsL.Léo​​ Ducas, E. W.​​​‌Eamonn W. Postlethwaite,​ L. N.Ludo N.​‌ Pulles and W. P.​​Wessel P. J. van​​​‌ Woerden. Hawk: Module​ LIP Makes Lattice Signatures​‌ Fast, Compact and Simple​​.ASIACRYPT (4)13794​​​‌Lecture Notes in Computer​ ScienceSpringer2022,​‌ 65--94back to text​​
  • 68 articleL.Léo​​​‌ Ducas. Provable lattice​ reduction of Zn with​‌ blocksize n/2.Designs,​​ Codes and CryptographyNovember​​​‌ 2023DOIback to​ text
  • 69 articleP.-A.​‌Pierre-Alain Fouque, P.​​Paul Kirchner, T.​​​‌Thomas Pornin and Y.​Yang Yu. BAT:​‌ Small and Fast KEM​​ over NTRU Lattices.​​​‌IACR Trans. Cryptogr. Hardw.​ Embed. Syst.20222​‌2022, 240--265URL:​​ https://doi.org/10.46586/tches.v2022.i2.240-265DOIback to​​​‌ text
  • 70 inproceedingsP.-A.​Pierre-Alain Fouque, P.​‌Paul Kirchner, M.​​Mehdi Tibouchi, A.​​​‌Alexandre Wallet and Y.​Yang Yu. Key​‌ Recovery from Gram-Schmidt Norm​​ Leakage in Hash-and-Sign Signatures​​​‌ over NTRU Lattices.​EUROCRYPT (3)12107Lecture​‌ Notes in Computer Science​​Springer2020, 34--63​​​‌back to text
  • 71​ inproceedingsP.-A.Pierre-Alain Fouque​‌, G.Gilles Macario-Rat​​ and J.Jacques Stern​​​‌. Key Recovery on​ Hidden Monomial Multivariate Schemes​‌.EUROCRYPT4965Lecture​​ Notes in Computer Science​​​‌Springer2008, 19--30​back to text
  • 72​‌ articleC.Craig Gidney​​ and M.Martin Eker\aa​​​‌. How to factor​ 2048 bit RSA integers​‌ in 8 hours using​​ 20 million noisy qubits​​​‌.Quantum52021​, 433URL: https://doi.org/10.22331/q-2021-04-15-433​‌DOIback to text​​
  • 73 articleC.Craig​​​‌ Gidney. How to​ factor 2048 bit RSA​‌ integers with less than​​ a million noisy qubits​​​‌.arXiv preprint arXiv:2505.15917​2025back to text​‌back to text
  • 74​​ inbookK.Keitaro Hashimoto​​​‌, S.Shuichi Katsumata​ and T.Thom Wiggers​‌. Bundled authenticated key​​ exchange: a concrete treatment​​​‌ of signal's handshake protocol​ and post-quantum security.​‌Proceedings of the 34th​​ USENIX Conference on Security​​​‌ SymposiumUSAUSENIX Association​2025back to text​‌
  • 75 inproceedingsA.Andreas​​ Hülsing, K.-C.Kai-Chun​​​‌ Ning, P.Peter​ Schwabe, F. J.​‌Fiona Johanna Weber and​​ P. R.Philip R.​​​‌ Zimmermann. Post-quantum WireGuard​.2021 IEEE Symposium​‌ on Security and Privacy​​ (SP)2021, 304-321​​DOIback to text​​​‌
  • 76 inproceedingsY.Yuval‌ Ishai, A.Amit‌​‌ Sahai and D. A.​​David A. Wagner.​​​‌ Private Circuits: Securing Hardware‌ against Probing Attacks.‌​‌CRYPTO2729Lecture Notes​​ in Computer ScienceSpringer​​​‌2003, 463--481back‌ to text
  • 77 article‌​‌M.Marc Kaplan,​​ G.Gaëtan Leurent,​​​‌ A.Anthony Leverrier and‌ M.María Naya-Plasencia.‌​‌ Quantum Differential and Linear​​ Cryptanalysis.IACR Trans.​​​‌ Symmetric Cryptol.20161‌2016, 71--94back‌​‌ to text
  • 78 inproceedings​​P.Paul Kirchner,​​​‌ T.Thomas Espitau and‌ P.-A.Pierre-Alain Fouque.‌​‌ Fast Reduction of Algebraic​​ Lattices over Cyclotomic Fields​​​‌.CRYPTO (2)12171‌Lecture Notes in Computer‌​‌ ScienceSpringer2020,​​ 155--185back to text​​​‌
  • 79 inproceedingsP.Paul‌ Kirchner and P.-A.Pierre-Alain‌​‌ Fouque. An Improved​​ BKW Algorithm for LWE​​​‌ with Applications to Cryptography‌ and Lattices.CRYPTO‌​‌ (1)9215Lecture Notes​​ in Computer ScienceSpringer​​​‌2015, 43--62back‌ to text
  • 80 inproceedings‌​‌P.Paul Kirchner and​​ P.-A.Pierre-Alain Fouque.​​​‌ Revisiting Lattice Attacks on‌ Overstretched NTRU Parameters.‌​‌EUROCRYPT (1)10210Lecture​​ Notes in Computer Science​​​‌2017, 3--26back‌ to text
  • 81 inproceedings‌​‌H.Hidenori Kuwakado and​​ M.Masakatu Morii.​​​‌ Quantum distinguisher between the‌ 3-round Feistel cipher and‌​‌ the random permutation.​​ISITIEEE2010,​​​‌ 2682--2685back to text‌back to text
  • 82‌​‌ inproceedingsH.Hidenori Kuwakado​​ and M.Masakatu Morii​​​‌. Security on the‌ quantum-type Even-Mansour cipher.‌​‌ISITAIEEE2012,​​ 312--316back to text​​​‌
  • 83 inproceedingsC.Changmin‌ Lee, A.Alice‌​‌ Pellet-Mary, D.Damien​​ Stehlé and A.Alexandre​​​‌ Wallet. An LLL‌ Algorithm for Module Lattices‌​‌.ASIACRYPT (2)11922​​Lecture Notes in Computer​​​‌ ScienceSpringer2019,‌ 59--90back to text‌​‌
  • 84 inproceedingsÉ.Éric​​ Levieil and P.-A.Pierre-Alain​​​‌ Fouque. An Improved‌ LPN Algorithm.SCN‌​‌4116Lecture Notes in​​ Computer ScienceSpringer2006​​​‌, 348--359back to‌ text
  • 85 miscMATZOV‌​‌. Report on the​​ Security of LWE: Improved​​​‌ Dual Lattice Attack.‌April 2022, URL:‌​‌ https://doi.org/10.5281/zenodo.6412487DOIback to​​ textback to text​​​‌back to textback‌ to textback to‌​‌ text
  • 86 inproceedingsV.​​Vincent Migliore, B.​​​‌Benoît Gérard, M.‌Mehdi Tibouchi and P.-A.‌​‌Pierre-Alain Fouque. Masking​​ Dilithium - Efficient Implementation​​​‌ and Side-Channel Evaluation.‌ACNS11464Lecture Notes‌​‌ in Computer ScienceSpringer​​2019, 344--362back​​​‌ to text
  • 87 article‌B.Brice Minaud,‌​‌ P.Patrick Derbez,​​ P.-A.Pierre-Alain Fouque and​​​‌ P.Pierre Karpman.‌ Key-Recovery Attacks on ASASA‌​‌.J. Cryptol.31​​32018, 845--884​​​‌back to text
  • 88‌ inproceedingsG.Guilhem Mureau‌​‌, A.Alice Pellet-Mary​​, G.Georgii Pliatsok​​​‌ and A.Alexandre Wallet‌. Cryptanalysis of Rank-2‌​‌ Module-LIP in Totally Real​​ Number Fields.EUROCRYPT​​​‌ (6)14656Lecture Notes‌ in Computer ScienceSpringer‌​‌2024, 226--255back​​ to textback to​​​‌ text
  • 89 articleG.‌Gwendal Patat, M.‌​‌Mohamed Sabt and P.-A.​​​‌Pierre-Alain Fouque. Your​ DRM Can Watch You​‌ Too: Exploring the Privacy​​ Implications of Browsers (mis)Implementations​​​‌ of Widevine EME.​Proc. Priv. Enhancing Technol.​‌202342023,​​ 306--321back to text​​​‌
  • 90 inproceedingsR.Rafael​ del Pino, S.​‌Shuichi Katsumata, M.​​Mary Maller, F.​​​‌Fabrice Mouhartem, T.​Thomas Prest and M.-J.​‌Markku-Juhani Saarinen. Threshold​​ Raccoon: Practical Threshold Signatures​​​‌ from Standard Lattice Assumptions​.Advances in Cryptology​‌ - EUROCRYPT 2024: 43rd​​ Annual International Conference on​​​‌ the Theory and Applications​ of Cryptographic Techniques, Zurich,​‌ Switzerland, May 26-30, 2024,​​ Proceedings, Part IIBerlin,​​​‌ HeidelbergZurich, SwitzerlandSpringer-Verlag​2024, 219–248URL:​‌ https://doi.org/10.1007/978-3-031-58723-8_8DOIback to​​ text
  • 91 inproceedingsA.​​​‌Amaury Pouly and Y.​Yixin Shen. Provable​‌ Dual Attacks on Learning​​ with Errors.EUROCRYPT​​​‌ (6)14656Lecture Notes​ in Computer ScienceSpringer​‌2024, 256--285back​​ to text
  • 92 inproceedings​​​‌B.Benjamin Pring.​ Exploiting Preprocessing for Quantum​‌ Search to Break Parameters​​ for MQ Cryptosystems.​​​‌WAIFI11321Lecture Notes​ in Computer ScienceSpringer​‌2018, 291--307back​​ to text
  • 93 inproceedings​​​‌E.Emmanuel Prouff and​ M.Matthieu Rivain.​‌ Masking against Side-Channel Attacks:​​ A Formal Security Proof​​​‌.EUROCRYPT7881Lecture​ Notes in Computer Science​‌Springer2013, 142--159​​back to text
  • 94​​​‌ inproceedingsM.Miruna Rosca​, D.Damien Stehlé​‌ and A.Alexandre Wallet​​. On the Ring-LWE​​​‌ and Polynomial-LWE Problems.​EUROCRYPT (1)10820Lecture​‌ Notes in Computer Science​​Springer2018, 146--173​​​‌back to text
  • 95​ inproceedingsA.André Schrottenloher​‌. Quantum Linear Key-Recovery​​ Attacks Using the QFT​​​‌.CRYPTO (5)14085​Lecture Notes in Computer​‌ ScienceSpringer2023,​​ 258--291back to text​​​‌
  • 96 techreportP.Peter​ Schwabe, R.Roberto​‌ Avanzi, J.Joppe​​ Bos, L.Léo​​​‌ Ducas, E.Eike​ Kiltz, T.Tancrède​‌ Lepoint, V.Vadim​​ Lyubashevsky, J. M.​​​‌John M. Schanck,​ G.Gregor Seiler and​‌ D.Damien Stehlé.​​ CRYSTALS-KYBER.available at​​​‌ https://csrc.nist.gov/projects/post-quantum-cryptography/round-3-submissionsNational Institute of​ Standards and Technology2020​‌back to textback​​ to text
  • 97 inproceedings​​​‌P. W.Peter W.​ Shor. Algorithms for​‌ Quantum Computation: Discrete Logarithms​​ and Factoring.FOCS​​​‌IEEE Computer Society1994​, 124--134back to​‌ text
  1. 1National Institute​​ for Standards and Technology​​​‌, a U.S. standardization​ agency whose cryptographic standards​‌ become de facto world​​ standards.