Skip to content

Robust Tangle Tree Displays All Nonsequential Separations

Claim/Theorem

Let \(M\) be a matroid or graph, and let \(\mathcal T\) be a tangle of order \(k\) satisfying the robustness hypothesis of Clark--Whittle.

Then there exists a tree decomposition of \(M\) that displays, up to the paper's natural tangle-equivalence relation, all \(k\)-separations of \(M\) that are non-trivial or nonsequential with respect to \(\mathcal T\).

Equivalently, once a high-order tangle is robust, its genuinely nontrivial low-order separations do not behave arbitrarily: modulo full-closure equivalence in the tangle, they are organized by a single tree.

For the Conjecture 3 frontier, this means the intrinsic gap can be reformulated as:

  1. prove the relevant high-order tangles are robust,
  2. show hardware-balanced low-rank cuts are nonsequential with respect to those tangles,
  3. exclude the remaining tree-or-flower configurations of such cuts.

Dependencies

  • [[tangle-order-equals-branchwidth.md]]

Conflicts/Gaps

  • The theorem requires the source paper's robustness hypothesis; current Conjecture 3 nodes do not yet prove that the relevant Quantum Tanner parity-check tangles satisfy it.
  • The conclusion only controls separations that are nonsequential with respect to the tangle, and only up to the paper's natural equivalence relation.
  • Therefore sequential low-order cuts, or many hardware cuts collapsing into one tangle-equivalence class or one flower, still remain possible at the present state of the graph.

Sources

  • 10.1016/j.jctb.2013.03.002