Figure 1
From: Recurrent quantum embedding neural network and its application in vulnerability detection
(a) The process of token 'NULL' being encoded into a quantum circuit. The \(|{\psi }_{1}\rangle\) obtained after rotation input layer is treated as a quantum one-hot vector. It further applies QEmbedding to obtain \(|{\psi }_{2}\rangle\) which can be treated as a quantum dense vector used as token representation. (b) Classical word embedding computation process. The one-hot vector is treated as a key to query value in the map of weights.
