Universal, Multipartite, and Shape-Dependent Entanglement

Universal entanglement data are what remain after local ultraviolet structure and allowed counterterms have been handled correctly. Multipartite and mixed-state measures then ask different questions of the same field state: total correlation, partial-transpose entanglement, canonical-purification correlation, or compatibility with an entropy cone. This chapter keeps those constructions separate while showing where geometry, operator content, and state enter.

Helpful background. Review UV divergences and the area law, mutual information, anomaly coefficients, and topological entanglement diagnostics as needed.

Geometry, state, and measure

A raw subregion entropy is dominated by local boundary terms. In a controlled family, anomaly-fixed logarithms, the odd-dimensional sphere constant, angle-dependent corner functions, mutual information at positive separation, or topological subtraction can remove part or all of that local ambiguity. Each construction has its own hypotheses: dimension, smoothness, state, algebra, regulator, and counterterm class.

Mixed-state measures add further choices. Negativity uses a partial transpose or graded analogue. Reflected entropy uses a canonical purification in a doubled algebra. Entropy cones collect all union entropies for a declared state class. None is defined merely by calling two regions “entangled.”

The first figure is the chapter dictionary. Begin at the central box, then choose the branch whose construction answers the physical question.

Region geometry, state, algebra, and regulator branch into universal shape terms, disjoint mutual information, negativity or reflected entropy, and multipartite entropy cones. — Universal geometric terms follow matched subtraction; disjoint mutual information follows positive separation; negativity and reflected entropy follow distinct mixed-state constructions; and entropy cones follow a specified multipartite state class. Schematic.

What the chapter establishes

The geometric sequence starts with Universal Terms and Entangling-Surface Geometry, develops perturbative Shape Dependence and Entanglement Variations, and treats Corners, Cusps, and Defect Data. Interval, Sphere, and Cylinder Entanglement in CFT then gives canonical conformal examples without rederiving specialist CFT data here.

The correlation and mixed-state sequence moves through Mutual Information for Disjoint Regions, Entanglement Negativity in QFT, and Reflected Entropy and Canonical Purifications. Multipartite Information and Entropy Cones distinguishes universal quantum inequalities from stabilizer and holographic restrictions.

The final sequence connects regulated Entanglement Spectra and Modular Spectral Data, Topological and Long-Range Entanglement Interfaces, and Finite-Temperature and Excited-State Entanglement. It ends with Choosing an Entanglement Measure for the Physical Question.

Conformal and replica-defect data are used here as inputs rather than rederived. Topological phase diagnosis, materials evidence, holographic geometric formulas, and special holographic entropy inequalities require their own model-specific analysis. Here the subject is the information measure itself, its continuum definition, and the matched comparison that licenses a QFT conclusion.

Entanglement constructions and their scientific domains
Question	Construction	Essential hypotheses	Boundary of the conclusion
Which shape coefficient is universal?	log term, sphere constant, corner function	dimension, state, smoothness or angle, counterterm class	Not every finite constant is universal
How are separated regions correlated?	mutual information and cross-ratio expansion	positive separation, algebras, state, OPE regime	Not a mixed-state entanglement measure
Is mixed-state entanglement present?	logarithmic negativity	partial transpose, grading, replica parity, cutoff	Zero need not imply separability
What does canonical doubling capture?	reflected entropy	algebra, support, canonical purification, split or regulator	Not a physical preparation or generic holographic area
Which multipartite vectors are allowed?	entropy cone for a state class	parties, unions, purifier, common subsystem prescription	Special cone inequalities are not generic QFT laws
Does a constant diagnose topology?	matched topological subtraction	gap, large regions, boundaries, interfaces, correlation length	Not a complete phase classification

A guided route

For geometry, read the first four pages in order and verify smooth, sharp, and conformal-map limits. For mixed-state correlation, begin with disjoint mutual information, then compare negativity and reflected entropy side by side. For multipartite structure, add entropy cones only after every union uses the same algebra and regulator. For continuum spectra or phase-related questions, read the spectrum and topological pages together with the relevant specialist treatments.

A recurring calculation compares the same free or conformal field with multiple regulators. The observable is not the raw cutoff-dependent entropy but a coefficient, balanced combination, or controlled low-lying scaling limit. Every fit includes geometry and regulator changes designed to falsify the proposed universality.

Universality has a domain

The second figure separates candidate universal information from the choices that remain. Its final box names common consequences of hiding those choices.

Potentially universal logarithms, corner functions, mutual information, and topological subtractions remain conditional on scheme, geometry, algebra, state, and measure; hidden choices produce false universality. — Matched subtraction can remove local cutoff terms, but counterterm class, geometry, interface, partial transpose, purification, continuation, ensemble, and state class still delimit the conclusion. Schematic.

For example, the sign $I_3\leq0$ is characteristic of the holographic entropy cone, not a generic quantum inequality. A nonzero topological subtraction can support intrinsic long-range order only after corner, boundary, interface, and correlation-length effects are controlled. A reflected-entropy calculation uses canonical auxiliary copies but does not assert that those copies are physical. These distinctions determine the scientific content.

Check your preparation

You are ready to use the chapter if you can:

Distinguish a universal logarithmic coefficient from a scheme-dependent constant.
Explain why a smooth shape kernel does not control a cusp without a new limit.
State the replica parity and fermionic convention needed for negativity.
Separate the generic quantum entropy cone from stabilizer or holographic cones.
List the finite-size, boundary, and correlation-length checks behind a topological subtraction.