Intrinsic Topological Entanglement Entropy and the Strong Subadditivity
Abstract: In $(2+1)d$ topological quantum field theory, topological entanglement entropy (TEE) can be computed using the replica and surgery methods. We classify all bipartitions on a torus and propose a general method for calculating their corresponding TEEs. For each bipartition, the TEEs for different ground states are bounded by a topological quantity, termed the intrinsic TEE, which depends solely on the number of entanglement interfaces $ \pi_{\partial A}$, $S_{\text{iTEE}}(A) = - \pi_{\partial A} \ln \mathcal{D}$ with $\mathcal{D}$ being the total quantum dimension. We derive a modified form of strong subadditivity (SSA) for the intrinsic TEE, with the modification depending on the genus $g_X$ of the subregions $X$, $S_{\text{iTEE}}(A) + S_{\text{iTEE}}(B) - S_{\text{iTEE}}(A\cup B) - S_{\text{iTEE}}(A\cap B) \geq -2\ln \mathcal{D} (g_A + g_B - g_{A\cup B} - g_{A\cap B})$. Additionally, we show that SSA for the full TEE holds when the intersection number between torus knots of the subregions is not equal to one. When the intersection number is one, the SSA condition is satisfied if and only if $\sum_a |\psi_a|2 (\ln S_{0a} - \ln |\psi_a|) + |S\psi_a|2 (\ln S_{0a} - \ln |S\psi_a|) \geq 2 \ln \mathcal{D}$, with $S$ being the modular $S$-matrix and $\psi_a$ being the probability amplitudes. This condition has been verified for unitary modular categories up to rank $11$, while counterexamples have been found in non-pseudo-unitary modular categories, such as the Yang-Lee anyon.
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Top Community Prompts
Collections
Sign up for free to add this paper to one or more collections.