On Entropy Bounds for Irrelevant Operators

Lucas Fernández-Sarmiento, Riccardo Penco and Rachel A Rosen · 2025

Read On Entropy Bounds for Irrelevant Operators by Lucas Fernández-Sarmiento, Riccardo Penco and Rachel A Rosen (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-14978.txt.

Introduction

Lucas Fernandez-Sarmiento, Riccardo Penco, and Rachel Rosen test in 2025 a guess that links entropy to small tweaks of a low energy theory. The guess says the top tweak that keeps a theory's rules must raise its entropy. Clues come from black hole heat math.

The team shows the guess equals a drop in heat grand value at fixed heat. They then check it on four known cases, from gold stone bits to a bent magnet model. Wide accord with the guess turns up. Cases where it should fail, such as rule breaking tweaks, are marked out as well.

Catalog record

CREATORLucas Fernández-Sarmiento, Riccardo Penco and Rachel A Rosen
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE55 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2508-14978

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

On Entropy Bounds for Irrelevant Operators Lucas Fernández-Sarmiento, Riccardo Penco and Rachel A Rosen Abstract Consistency constraints for low-energy theories, especially those lacking Lorentz invariance, have recently garnered attention. Building on results from black hole thermodynamics, we investigate the conjecture that leading irrelevant deformations of a conformal field theory in the infrared must increase the system’s entropy. We show that this entropy-positivity conjecture is equivalent to a decrease in the thermal grand potential at a fixed temperature. We then evaluate this proposal against various known positivity bounds and other physical constraints on effective theories: for U ⁡ ( 1 ) U(1) Goldstone bosons with a quartic self-interaction at (non-)zero chemical potential, for the Euler-Heisenberg model, for the O ⁡ ( N ) O(N) nonlinear sigma model in ( 2 + 1 ) ​ D (2+1)D , and for T ​ T ¯ T\bar{T} deformations of the 2D Ising CFT. We find broad agreement with the entropy-positivity conjecture and we discuss test cases where the conjecture cannot be applied. 1 Introduction The standard assumption of Effective Field Theory (EFT) is that low energy theories can be constructed in a model independent way based solely on the input of the particle content and the symmetry of the low energy theory. From this perspective, one can remain entirely agnostic about the microscopic physics. Wilson coefficients of a given theory are then taken to be fixed by experiment or observation rather than through theoretical arguments. However, there has been great recent interest in using the principles of theoretical consistency such as causality or unitarity to constrain Wilson coefficients of low energy effective theories (see, e.g., deRham:2022hpx and references therein). Consistency constraints for theories without Lorentz invariance are notoriously hard to obtain. For instance, in Adams:2006sv positivity bounds are derived for Lorentz invariant theories, but in general, one cannot rely on many of the analytic properties that Lorentz invariance confers when dealing with systems where this symmetry is non-linearly realized. Indeed, recent work has shown non-analyticities in the dispersion relation as a byproduct of Lorentz symmetry breaking, leading to branch cuts in unphysical regions of kinematic space Hui:2023pxc ; Creminelli:2023kze . Nevertheless, progress has been made along a variety of axes (see, e.g., Creminelli:2022onn ; Creminelli:2024lhd ). In Cheung:2018cwt an entropic argument was invoked to provide bounds on the coefficients of higher-dimensional operators that contribute to the entropy of a black hole. There it was shown that, for a thermodynamically stable black hole and a theory in which the dominant contributions to higher dimension operators are the tree level contributions generated by massive fields that have been integrated out, the following inequality holds at fixed temperature T=1/\beta : F(\beta)<F_{0}(\beta)\,, (1) where F(\beta) is the free energy of a theory that contains microscopic degrees of freedom that have been frozen out in the computation of F_{0}(\beta) . For instance, F_{0} can be taken to be the free energy of a Reissner-Nordström black hole with a pure Einstein-Maxwell Lagrangian, while F includes higher order operators. More generally F_{0} can refer to a theory in which the cutoff has been taken to infinity. We emphasize that the proof of ( 1 ) relies on the fact that the dominant contribution to black hole thermodynamics comes from the saddle point, and on the assumption of thermodynamical stability, i.e. that the saddle point is a minimum of the Euclidean action. The authors of Cheung:2018cwt then show that the inequality ( 1 ) is equivalent to an inequality of the entropies at fixed black hole mass M : F(\beta)<F_{0}(\beta)\implies S(M)>S_{0}(M)\,. (2) These entropy inequalities give new positivity bounds on the coefficients of higher-dimension operators in gravitational theories. In this paper, we aim to exploit the fact that the entropy constraint \Delta S>0 can be applied to theories without Lorentz invariance, unlike many traditional positivity bounds. While we cannot prove F(\beta)<F_{0}(\beta) away from the saddle point approximation, we conjecture that it must be true for irrelevant deformations of an infrared conformal field theory even when the leading contribution to the free energy comes from loops. This conjecture is a weaker statement than the claim that the thermal free energy should be monotonic along the RG flow Appelquist:1999hr ; Appelquist:1999vs ; Periwal:1994im , for which there are known counterexamples in the literature Sachdev:1993pr ; Klebanov:2011gs . We will check the validity of our conjecture in a variety of known systems. Our assumption has physical justification. Namely, irrelevant corrections to higher dimension operators encode the presence of microscopic degrees of freedom that have been integrated out, and which presumably would give rise to a larger number of microscopic configurations corresponding to the same macroscopic state Kats:2006xp . Thus, we would expect a theory with these operators to have a greater entropy than a theory in which the cutoff (i.e., the mass of the heavy fields) has been taken to infinity and the heavy degrees of freedom are decoupled. Note that the contribution to the entropy coming from the dispersion relation of the heavy modes of mass M will be suppressed as e^{-\beta M} compared to the contributions from irrelevant corrections to the low-energy Lagrangian, which will be polynomial in the low temperature regime \beta M\gg 1 . In what follows, we show in full generality that, whenever an inequality exists for the grand potential energy \Delta\Omega(\beta,\mu)<0 , it implies an equivalent inequality for the entropy as a function of the extensive variables, \Delta S(E,Q)>0 , without any saddle point approximation or assumption of Lorentz invariance. We will then apply the bound \Delta S>0 to multiple systems for which the leading thermodynamic behavior doesn’t come from the saddle point. We will find that the bounds from the entropy inequality pass some nontrivial consistency checks and are compatible with known positivity arguments in all the examples we consider. The outline of this paper is as follows. In section 2 we will construct a general equivalence between a negative shift in grand potential energy \Omega(\beta,\mu) and a positive shift in entropy S(E,Q) . In section 3 we will briefly discuss entropy considerations for a non-interacting theory. In section 4 we will consider several applications of our entropy bounds, to theories both with and without Lorentz invariance. In section 5 , we explore cases that fall outside the scope of our bounds, where these results may not apply. We end with some discussion in section 6 . Conventions and Notation: We work in the mostly plus metric signature (-,+,+,+) , we adopt natural units such that c=\hbar=k_{B}=1 , and we restrict ourselves to 4 spacetime dimensions unless otherwise indicated. 2 Entropy and Grand Potential Energy In Cheung:2018cwt , it was shown that a negative change in the free energy F at fixed temperature \Delta F(\beta)<0 is equivalent to a positive change in the entropy S at fixed energy \Delta S(E)>0 . Here we generalize this result by considering systems at finite chemical potential \mu . Thus, instead of the Helmholtz free energy F , we will be using the grand potential \Omega\equiv-T\ln{\cal Z}=F-\mu\,Q . A known subtlety with the grand potential (and free energy) in quantum field theory is its sensitivity to UV divergences which are independent of temperature, such as the zero-point energy. To construct an argument

Read the complete text