A theoretical study finds that a proposed rule for how quickly masses should fall near distant limits of a field theory is not universal across the examples it tests. The Sharpened Distance Conjecture, or SharpDC, holds at leading order in a large-N set of supersymmetric gauge theories. But the analysis also reports violations in a non-supersymmetric one-loop setting, a Class S example at N=2, and an orthosymplectic circular-quiver family at every finite N. The work is an arXiv preprint based on analytical calculations of field-theory quantities and holographic geometries.
The paper tests the CFT versions of the Sharpened Distance Conjecture and the related Refined Distance Conjecture, or RDC, in AdS5 settings through the conformal field theory dual. SharpDC is expressed here as a lower limit on the exponential rate at which a tower of higher-spin states becomes light, while the RDC is treated as a limit on field range and moduli-space distance. The method follows weak-coupling limits, identifies the towers, and computes that rate from CFT data.
The positive result has a precise boundary
The clearest positive result comes from large-N Lagrangian gauge theories with a finite number M of gauge factors and a vanishing one-loop beta-function. Here N is the gauge-theory rank taken large, while M remains finite. In the leading-order supersymmetric calculation, the mass-decay rate is at least one divided by the square root of two. The bound is saturated when fundamental chiral matter does not grow with N.
Within that same finite-factor setting, the SharpDC test can be recast as a comparison between two quantities associated with a four-dimensional conformal theory. At leading order, it requires c, the conformal central charge, to be at least a, the conformal anomaly coefficient. The equivalence narrows what the positive result means: it is a leading large-N statement about the specified supersymmetric gauge-theory family.
Other families expose cracks
The non-supersymmetric calculation produces a different answer. Under fixed-factor, weak-coupling Lagrangian assumptions, a theory with a vanishing one-loop beta-function can violate SharpDC. The paper gives 11 Weyl fermions in an index-two representation of SU(N) as a saturating example of the less-strict rate bound found in that analysis. This result is tentative in scope because vanishing of the one-loop beta-function is necessary but generally not sufficient to establish exact conformality.
Class S provides another stress test. These are four-dimensional N=2 superconformal field theories obtained from six-dimensional N=(2,0) theories on surfaces with genus greater than 1 and no punctures. In the reported Class S example, the case labeled N=2 has a rate equal to the square root of the fraction 13 over 27, below one divided by the square root of two, so the CFT SharpDC is violated. The paper reports that this is the only value in the stated example where the comparator is crossed.
A separate orthosymplectic circular-quiver family violates SharpDC for every finite N, but the violation becomes subleading as N grows. The authors interpret parametrically large M in related constructions as partial decompactification, meaning that part of the internal geometry becomes much larger than the AdS space. Together, these examples make the size of the limit matter: a result that emerges at large N need not describe each finite-N member.
The unresolved geometry behind the RDC
The RDC analysis moves from rates to geometry. For the Class S theories, the relevant moduli space is the Weil-Petersson space associated with the surfaces. In five-dimensional Planck units, the reported AdS moduli-space metric is the Weil-Petersson metric divided by 2 pi times the genus minus one. That identification turns the AdS field-range question into a question about the diameter of the Weil-Petersson space.
Near nodal degenerations, the paper estimates two shrinking distance scales. The first correction region begins at a Weil-Petersson distance that decreases as the negative one-sixth power of N. A smaller threshold, decreasing as the negative one-half power of N, is where the string tension reaches the AdS scale and no Einstein description remains. These are asymptotic estimates within the stated holographic approximation, and finite-N corrections to the moduli-space metric are not analyzed.
At large N, the AdS RDC becomes a proposed upper bound on the Weil-Petersson diameter. In ordinary terms, the greatest separation between points in that space should grow no faster than square-root order in the genus. The mathematical bounds summarized by the paper leave that proposal unresolved: as genus grows, the reported lower growth is of square-root order, while the upper growth is of square-root-times-logarithm order. For the RDC implication to hold, the diameter would have to saturate the lower scaling.
The preprint therefore delivers a mixed result. The leading large-N supersymmetric family meets the SharpDC rate bound, while the non-supersymmetric, finite-N, and Class S examples show that the reported success does not cover every family examined. The work also turns the AdS RDC into a concrete mathematical question about Weil-Petersson diameter, but the proposed upper bound remains unproven.
Paper data and sources
Original title: Stress-Testing Swampland Bounds with Class S Theories
Authors: Gabriel Fenati, Miguel Montero, Irene Valenzuela
Journal/Repository: arXiv
Status: Preprint, not yet peer-reviewed
First online: 2026-08-28
DOI: Not available
Original paper · Full text