Exact endpoints in the proof of combinatorial invariance
The complete schedule has two actual sheaf-character endpoints. The final state uses stalks, while the initial state uses full outgoing kernels and their duality description. Scaling expresses these endpoints in the form needed for the normalized R-matrix comparison.
After the R-entries of the source and target intervals have been identified, character reciprocity makes the transfer-product comparison an exact endpoint equality. This is the role of the endpoint-equality theorem collected under this keyword. Positivity then forces each intermediate step to be exact, and the image-equality criterion supplies the module identities used in reverse freeness.
This endpoint step belongs inside the simultaneous R-invariance and suffix-freeness induction. The canonical KL polynomial is identified afterward by its own reciprocity and degree uniqueness. The final theorem combines that finite-generator result with finite-support interval models. The route below follows the endpoint data into the whole conclusion.
Reading sequence
- The intrinsic graded-rank polynomial of a free module — Define the actual sheaf character.
- Kernel Hilbert values and normalized states — Define the numerical state and its scale.
- Normalized empty and full kernels from the actual character — Identify the actual initial and final values.
- R-reciprocity for the actual boundary-sheaf character — Read the source reciprocity relation.
- R-matrix equality makes the comparison endpoints exact — Apply equality of R-data to the endpoints.
- Endpoint equality forces equality at every transfer step — Propagate endpoint equality back to each step.
- Simultaneous R-tilde invariance and terminal-kernel freeness — Place the step in the simultaneous induction.
- KL invariance for finite-generator Coxeter systems — Identify the canonical KL polynomials.
- Combinatorial invariance of Kazhdan–Lusztig polynomials — Read the final unrestricted statement.