From 28a7535e6acb4d2edfe244c33d34a96892e7fae6 Mon Sep 17 00:00:00 2001 From: Brian W Bush Date: Fri, 27 Dec 2024 11:33:42 -0700 Subject: [PATCH] Added reference to Leios paper's VRF, BLS, and voting. --- docs/technical-report-1.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/technical-report-1.md b/docs/technical-report-1.md index cb11bfc9..4b9bf828 100644 --- a/docs/technical-report-1.md +++ b/docs/technical-report-1.md @@ -81,7 +81,7 @@ $$ \phi_f(\sigma) = \psi_{f^\prime}(\sigma \cdot S) = 1 - (1 - f)^\sigma $$ -Sortition reuses the Praos VRF, but prepending a string such as `Leios-IB`, `Leios-EB`, or `Leios-Vote` to the bytes being hashed, in order to remove correlations between the uses of the VRF. The resulting VRF value `v` is compared to the probability (in the case of IBs and EBs) or the cumulative probability (in the case of votes) for winning the sortition lottery. This is equivalent to the eligibility checks in the Leios paper. +Sortition reuses the Praos VRF, but prepending a string such as `Leios-IB`, `Leios-EB`, or `Leios-Vote` to the bytes being hashed, in order to remove correlations between the uses of the VRF. Figures 6, 7, and 8 of the Leios paper describe the eligibility computations. In essence, the resulting eligibility value $v$ is compared to the probability (in the case of IBs and EBs) or the cumulative probability (in the case of votes) for winning the sortition lottery. ### Advantage gained by splitting stake among nodes