Scheinerman–Ullman fractional Hamiltonicity conjecture and Devriendt's resistance-positivity conjecture

Every finite graph G=(V,E)G=(V,E) that is 22-tough is fractionally Hamiltonian. Here, GG is 22-tough if, for every vertex set S⊆VS\subseteq V with c(G−S)>1c(G-S)>1, one has ∣S∣≥2c(G−S)|S|\ge 2c(G-S), where c(G−S)c(G-S) denotes the number of connected components of G−SG-S. The graph GG is fractionally Hamiltonian if there exists a function x:E→[0,1]x:E\to[0,1] such that ∑e∈Exe=∣V∣\sum_{e\in E}x_e=|V| and, for every nonempty proper subset S⊂VS\subset V, ∑e∈δ(S)xe≥2\sum_{e\in\delta(S)}x_e\ge 2, where δ(S)\delta(S) is the edge cut between SS and V∖SV\setminus S.

References

Progress summary

Refreshed
Claimed progress

A new paper proves the conjecture for graphs at least five times tougher than required, while the broader resistance claim has reportedly been disproved.

The Scheinerman–Ullman conjecture asks whether every 22-tough graph is fractionally Hamiltonian; Devriendt asked whether every 11-tough graph is resistance positive.

Known results

  • Every 1010-tough chordal graph with at least 33 vertices is Hamilton-connected (2015).

September 2026 developments

  • Toughness Bounds for Fractional Hamiltonicity and Resistance Positivity claims that every connected non-fractionally-Hamiltonian graph has a non-Hamiltonian chordal spanning supergraph; combined with the chordal theorem, this yields fractional Hamiltonicity for every 1010-tough graph and resistance positivity in that class.
  • A July 2026 paper claims Devriendt’s broader resistance conjecture is false: for every n≥11n \ge 11, there is an nn-vertex 11-tough graph that is not resistance nonnegative.

Current status (as of September 2026): Fractional Hamiltonicity is established for 1010-tough graphs, but the 22-tough conjecture remains open; the broad 11-tough resistance-positivity conjecture is claimed false, while positivity for the 1010-tough class is claimed.

Sources

Solutions 0

No solutions have been posted yet.