Update 2024-01-31-Sumcheck.md

Author: RisenCrypto

_posts/2024-01-31-Sumcheck.md

@@ -550,8 +550,8 @@ In all the steps where $\mathcal P$ has to send a univariate $g_i$, she instead
 
 Note that the above program has many calculations which are redone repeatedly & hence can be optimized further by dynamic programming.
 
-In the final step of the protocol, $\mathcal V$ has to compute $g(x_1,x_2,x_3, x_4, x_5, x_6)$ for random values. 
-The book says pick $r_1, r_2, r_3 \in \mathbb F^{\log n} \times \mathbb F^{\log n} \times \mathbb F^{\log n}$. In our case, $\log n = v = 2$, so we have to pick 6 random numbers $r_1$ will be the vector of the first 2 random numbers, $r_2$ the second two and so on. Computing $g$  can be done in time linear to the size of the input Matrix.
+In the final step of the protocol, $\mathcal V$ has to compute $g$ with random values. 
+The book says pick $r_1, r_2, r_3 \in \mathbb F^{\log n} \times \mathbb F^{\log n} \times \mathbb F^{\log n}$. In our case, $\log n = v = 2$, so we have to pick 6 random numbers. Computing $g$ can be done in time linear to the size of the input Matrix.
 
 **This post is based on Justin Thaler's Book [Proofs, Arguments, and Zero-Knowledge
 ](https://people.cs.georgetown.edu/jthaler/ProofsArgsAndZK.html)**