Bond-weighting method for the Grassmann tensor renormalization group

A bstract Recently, the tensor network description with bond weights on its edges has been proposed as a novel improvement for renormalization group algorithm. The weight is controlled by single hyperparameter, whose optimal value estimated in original work via numerical computation of two-dimensional critical Ising model. We develop this bond-weighted algorithm to make it applicable fermionic system, benchmarking massless Wilson fermion. show that accuracy fixed dimension improved also system and provide evidence choice hyperparameter not affected whether bosonic or fermionic. In addition, monitoring singular spectrum, we find scale-invariant structure renormalized Grassmann successfully kept bond-weighting technique.