TSThomas Sirvent

Thomas Sirvent

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Thomas Sirvent has a total of 10 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

    • 12 December 2015
    • Published by HAL in HAL
    Open Access

    Effectiveness of Synthesis in Concolic Deobfuscation

    by Fabrizio Biondi, Sébastien Josse, Axel Legay, Thomas Sirvent

    The obfuscation of conditional statements is a simple and efficient way to disturb the identification of the control flow graph of a program. Mixed Boolean arithmetics (MBA) techniques provide concrete ways to achieve this obfuscation of conditional statements. In this work, we study the effectiveness of automated deobfuscation of MBA obfuscation, using algebraic, SMT-based and synthesis-based techniques. We experimentally ascertain the practical feasibility of MBA obfuscation. We study using SMT-based approaches with different state-of-the-art SMT solvers to counteract MBA obfuscation, and we show how the deobfuscation complexity can be greatly reduced by algebraic simplification. We also consider synthesis-based deobfuscation and find it to be more effective than SMT-based deobfuscation. We discuss complexity and limits of all methods, and conclude that MBA obfuscation is not effective enough to be considered a reliable method for control flow or white-box obfuscation.

  • Open Access

    Privacy-Preserving Reputation Mechanism: A Usable Solution Handling Negative Ratings

    by Paul Lajoie-Mazenc, Emmanuelle Anceaume, Gilles Guette, Thomas Sirvent, Valérie Viet Triem Tong


    • 23 January 2015
    • Published by HAL
    Open Access

    Efficient Distributed Privacy-Preserving Reputation Mechanism Handling Non-Monotonic Ratings

    by Paul Lajoie-Mazenc, Emmanuelle Anceaume, Gilles Guette, Thomas Sirvent, Valérie Viet, Triem Tong

    —Open and large-scale systems do not encourage their users to behave trustworthily, which may entail non-negligible risks when interacting with unknown users, for instance when buying an item on an e-commerce platform. Reputation mech-anisms reduce these risks by associating a reputation score to each user, summarizing their past behavior. To be useful to users, reputation mechanisms need to guarantee two main properties: the non-monotonicity of reputation scores, in order to exactly reflect the users' behavior, and the privacy of their users, so that the history of their transactions is not publicly available. We propose a distributed privacy-preserving reputation mechanism handling non-monotonic ratings. Our proposition relies on two distinct distributed third parties and on cryptographic tools, including zero-knowledge proofs of knowledge, anonymous proxy signatures, and verifiable secret sharing. We show that this proposal is computationally efficient, and thus practical. To the best of our knowledge, this solution is the first one that preserves users' privacy and handles both positive and negative ratings without relying on a central authority.

    • 23 January 2015
    • Published by HAL
    Open Access

    Efficient Distributed Privacy-Preserving Reputation Mechanism Handling Non-Monotonic Ratings

    by Paul Lajoie-Mazenc, Emmanuelle Anceaume, Gilles Guette, Thomas Sirvent, Valérie Viet, Triem Tong

    —Open and large-scale systems do not encourage their users to behave trustworthily, which may entail non-negligible risks when interacting with unknown users, for instance when buying an item on an e-commerce platform. Reputation mech-anisms reduce these risks by associating a reputation score to each user, summarizing their past behavior. To be useful to users, reputation mechanisms need to guarantee two main properties: the non-monotonicity of reputation scores, in order to exactly reflect the users' behavior, and the privacy of their users, so that the history of their transactions is not publicly available. We propose a distributed privacy-preserving reputation mechanism handling non-monotonic ratings. Our proposition relies on two distinct distributed third parties and on cryptographic tools, including zero-knowledge proofs of knowledge, anonymous proxy signatures, and verifiable secret sharing. We show that this proposal is computationally efficient, and thus practical. To the best of our knowledge, this solution is the first one that preserves users' privacy and handles both positive and negative ratings without relying on a central authority.

    • 23 January 2015
    • Published by HAL in HAL
    Open Access

    Efficient Distributed Privacy-Preserving Reputation Mechanism Handling Non-Monotonic Ratings

    by Paul Lajoie-Mazenc, Emmanuelle Anceaume, Gilles Guette, Thomas Sirvent, Valérie Viet, Triem Tong

    —Open and large-scale systems do not encourage their users to behave trustworthily, which may entail non-negligible risks when interacting with unknown users, for instance when buying an item on an e-commerce platform. Reputation mech-anisms reduce these risks by associating a reputation score to each user, summarizing their past behavior. To be useful to users, reputation mechanisms need to guarantee two main properties: the non-monotonicity of reputation scores, in order to exactly reflect the users' behavior, and the privacy of their users, so that the history of their transactions is not publicly available. We propose a distributed privacy-preserving reputation mechanism handling non-monotonic ratings. Our proposition relies on two distinct distributed third parties and on cryptographic tools, including zero-knowledge proofs of knowledge, anonymous proxy signatures, and verifiable secret sharing. We show that this proposal is computationally efficient, and thus practical. To the best of our knowledge, this solution is the first one that preserves users' privacy and handles both positive and negative ratings without relying on a central authority.