 Hi everybody! I'm going to present you a giant war entitled syndrome liquidation in the head shorted signatures from zero-knowledge proofs. In this war, we propose a new codebase signature scheme built from zero-knowledge proof. To design this signature, we chose to focus on the syndrome decoding problem for which cryptanalysis is stable for many years. Then, we used the MPC in the head paradigm to build a zero-knowledge proof of knowledge for this problem. This paradigm provides a generic method to build zero-knowledge protocols using techniques from secure multi-parted computation. To apply the MPC in the head paradigm, we need to have an efficient MPC protocol that checks that a vector is a solution of the syndrome decoding instance by checking those two relations. The linear relation is easy to check with MPC, but it is much harder for the constraint on the having weight. For this reason, we replaced the second constraint with an equivalent one involving polynomials. This new constraint consists to prove the existence of a polynomial Q such that f divides the product sq, where s is a polynomial for which the evaluations and some product points give the coordinates of the test vector, and f is the smallest polynomial for which have for roots all those evaluation points. In practice, such polynomial Q can be easily built from the tested vector. We simply need to take a unique polynomial of degree w so that the non-zero position of the tested vector in coding has roots. Now that we have the exact statement we want to check, we can use the usual techniques to build the wanted MPC protocol. Then we convert this protocol into an interactive zero-knowledge proof of knowledge using the MPC in the head paradigm. Finally, we apply the Fiat-Scharmier transformation to obtain the signature scheme. We selected three parameter sets targeting a security of 128 bits. The two first sets rely on the syndrome decoding instance on the binary field, while the third one relies on the syndrome decoding instance on a larger field. For each parameter set, we proposed two trailers. The first one lowers the communication cost, and the second one lowers the computational cost. Let us compare our scheme with the other code-based signature built from zero-knowledge proofs. We can observe that the obtained signature sizes are less than for all the former schemes. We have the first such scheme, which can produce signature size along the symbolic cap of 10 kilobytes. Now, let us compare our scheme with all the other code-based signatures. We are far to have the shortest signature. However, our scheme outperforms all the signatures when considering the sum of the size of the signature and the public. To conclude, we proposed in this work a new code-based signature relying on the syndrome decoding problem for random linear codes. And this signature has competitive performance regarding the current state of the art. Thank you for your attention and see you on the 16th of August for the complete talk.