Center for Theoretical Physics

Scale separation is an important physical principle that has previously enabled algorithmic advances such as multigrid solvers. Previous work on normalizing flows has been able to utilize scale separation in the context of scalar field theories, but the principle has been largely unexploited in the context of gauge theories. This work gives an overview of a new method for generating gauge fields using hierarchical normalizing flow models. This method builds gauge fields from the outside in, allowing different parts of the model to focus on different scales of the problem. Numerical results are presented for \(U(1)\) and \(SU(3)\) gauge theories in 2, 3, and 4 spacetime dimensions.

We study the general framework of effective theories obtained via exact path integration of a complete theory over some sector of its degrees of freedom. Theories constructed this way contain multi-integrals which couple fields arbitrarily far apart, and in certain settings even on path-disconnected components of the space. These are not just entanglement, but genuine non-local interactions that we dub quantum wormholes. Any state the path integral of such an effective theory prepares is shown to be a partial trace of a state of the complete theory over the integrated-out sector. The resulting reduced density operator is generally mixed due to bra-ket wormholes. An infinite family of ensembles of pure states of the complete theory giving the same effective state is identified. These allow one to equivalently interpret any effective state as being prepared by an ensemble of theories. When computing entropic quantities, bra-ket wormholes give rise to replica wormholes. This causes replica path integrals for the effective theory to not factorize even when the underlying manifold does, as expected from mixing. In contrast, effective theories obtained by derivative expansions have no quantum wormholes and prepare pure states. There exist operators in the algebra of effective theories which can distinguish mixed from pure states, implying a breakdown of non-exact effective theories for sufficiently complex observables. This framework unifies and provides new insights into much of the phenomena observed in quantum gravity, including the interplay between wormholes and unitarity, the breakdown of bulk effective theory, the factorization puzzle, state ensembles, theory ensembles, quantum error correction, and baby universes. Some interesting lessons are drawn accounting also for characteristic aspects of gravity concerning IR/UV mixing and Kaluza-Klein reductions.

The transverse momentum-dependent fragmentation functions (TMD FFs) of heavy (bottom and charm) quarks, which we recently introduced, are universal building blocks that enter predictions for a large number of observables involving final-state heavy quarks or hadrons. They enable the extension of fixed-order subtraction schemes to quasi-collinear limits, and are of particular interest in their own right as probes of the nonperturbative dynamics of hadronization. In this paper we calculate all TMD FFs involving heavy quarks and the associated TMD matrix element in heavy-quark effective theory (HQET) to next-to-leading order in the strong interaction. Our results confirm the renormalization properties, large-mass, and small-mass consistency relations predicted in our earlier work. We also derive and confirm a prediction for the large-\(z\) behavior of the heavy-quark TMD FF by extending, for the first time, the formalism of joint resummation to capture quark mass effects in heavy-quark fragmentation. Our final results in position space agree with those of a recent calculation by another group that used a highly orthogonal organization of singularities in the intermediate momentum-space steps, providing a strong independent cross check. As an immediate application, we present the complete quark mass dependence of the energy-energy correlator (EEC) in the back-to-back limit at \(\mathcal{O}(\alpha_s)\).

We study future lepton collider prospects for testing predictive models of leptophilic dark matter candidates with a thermal origin. We calculate experimental milestones for testing the parameter space compatible with freeze-out and the associated collider signals at past, present, and future facilities. This analysis places new limits on such models by leveraging the utility of lepton colliders. At \(e^+e^-\) machines, we make projections using precision \(Z\)-pole observables from \(e^+e^-\to l^+l^- + \) missing energy signatures at LEP and future projections for FCC-ee in these channels. Additionally, a muon collider could also probe new thermal relic parameter space in this scenario via \(\mu^+\mu^- \to X + \) missing energy where \(X\) is any easy identifiable SM object. Collectively, these processes can probe much all of the parameter space for which DM direct annihilation to \(l^+l^-\) yields the observed relic density in Higgs-like models with mass-proportional couplings to charged leptons.