statistical physicists

Abstract: We give algorithms for approximating the partition function $Z_N$ of the Sherrington--Kirkpatrick (SK) model and finite mixed Ising $p$-spin models with $p\ge 3$. The algorithms apply at a fixed margin below the second-moment threshold $β_{\mathrm f}$ with homogeneous weak external fields $|h|\leqρ_0N^{-1/4}$, for fixed $ρ_0>0$. For any accuracy $\varepsilon\in(0,1)$ and failure parameter $δ\in(0,1)$, they approximate $\log Z_N$ to additive error $\varepsilon$. The number of exact operations is polynomial jointly in $N$, $\varepsilon^{-1}$, and $δ^{-1}$. The disorder variables have i.i.d symmetric distribution with unit variance. The failure probability is at most $δ$ plus an explicit disorder tail bound. This tail bound tends to zero under a finite fourth moment condition. Our approach builds on the cluster expansions of Aizenman--Lebowitz--Ruelle~\cite{ALR87} and Dey--Wu~\cites{DW23,DW26}. For higher-order mixtures, we retain hypergraphs using few vertices and evaluate their total weight deterministically through local spin averages and inclusion--exclusion. For SK, we impose cutoffs on the number of vertices of degree at least three and on the total number of edges. We estimate the retained sum using randomized color coding~\cite{AYZ95}. A lower tail control based on reverse hypercontractivity~\cite{MOS13} converts the mean-square approximation of the normalized partition function into logarithmic accuracy.

Abstract: Accelerated dynamics methods, such as parallel replica dynamics, require estimates of the decorrelation time, after which a process trapped in one of its metastable states is distributed according to the quasi-stationary distribution (QSD). In a method proposed by Binder, Lelièvre and Simpson, decorrelation times are defined by coupling a particle approximation of the non-linear conditioned probability flow of the process with MCMC-inspired diagnostics. We explore an alternative to MCMC diagnostics, based on coarse-graining of non-linear probability flows and simulation-based inference. We first develop a coarse-graining framework for killed Markov jump processes conditioned on survival, proving uniform-in-time, non-asymptotic estimates approximating the non-linear probability flow of the true process by the non-linear flow of a surrogate process, defined with a collective variable. The framework does not assume reversibility of the true process, and also gives bounds which scale explicitly in an adiabatic regime. Motivated by these results, we then construct a quasi-stationarity diagnostic using simulation-based inference: a recurrent neural network (RNN), trained on particle approximations of non-linear effective flows, estimates posterior probabilities of decorrelation. The learned diagnostic is then transferred to collective-variable traces of high-dimensional particle approximations. We investigate the behavior and robustness of this diagnostic on a set of numerical experiments, comparing it to the MCMC diagnostic of Binder et al.\ on both synthetic examples and a molecular system, in and out of its training prior distribution. When its prior assumptions hold, the RNN diagnostic outperforms the MCMC diagnostic in ROC analysis, is less conservative, fails less often, and has an interpretable decision threshold.

Abstract: We establish a reference modeling framework for random probability distributions subject to constraints. Generalizing the hyperensemble approach of Crooks, we formulate Jaynesian priors of maximal uncertainty over the unit simplex of categorical distributions. This framework is entirely general, but we primarily investigate models with constraints on physically relevant quantities, namely, average energy and Shannon entropy. A constraint on the exact value of average energy uniquely determines a Boltzmann-Gibbs equilibrium, but also defines a hyperplanar subset of the probability simplex consisting of all distributions with that value of average energy. The associated hyperensemble has a uniform probability density across all such distributions, regardless of distance to equilibrium. Such reference models make predictions of generic nonequilibrium properties such as the mean and modal microstate distributions and the expected free energy of a random distribution, which we examine analytically and via exact sampling algorithms. Moreover, introducing constraints on Shannon entropy permits parametric access to tunably nonequilibrium regimes. The approach follows conventional Jaynesian modeling logic and is compatible with standard Bayesian inference, and reduces to Boltzmann-Gibbs statistics when Shannon entropy is maximized. We examine the physical predictions of this scheme, and demonstrate its wide compatibility with other approaches to nonequilibrium modeling. We illustrate the framework in three scenarios: (i) comparison to superstatistics quasi-equilibrium modeling, (ii) generation of random nonequilibrium steady states, and (iii) Bayesian inference under a prior hyperensemble. We highlight numerous prospects for theoretical extensions and empirical application.

Abstract: We study Glauber dynamics for the Ising model on random $d$-regular graphs with $n$ vertices at low temperature, in the regime $β\geq C/d$ where $C,d$ are sufficiently large constants. In this regime, the full mixing time is exponential in $n$, because of the bottleneck between the two dominant phases: configurations with a majority of plus spins and configurations with a majority of minus spins. We show that, for sufficiently large $d$, if the dynamics is initialized with an excess of at least $ε_{C,d} n$ plus spins over minus spins, then it quasi-equilibrates in time $O(\log n)$ to the Gibbs measure conditioned on positive magnetization. In addition, the required bias $ε_{C,d}$ can be chosen so that $ε_{C,d}\to 0$ as $d,C\to\infty$. Previous bound of \cite{GSS25+} in the Ising and large-$d$ setting required $β\gtrsim (\log d)/d$. A key difficulty is that, even from the all-plus initialization, one must control the propagation of negative information in spacetime in a low-temperature model with strong local correlations. We address this by developing a new bootstrap approach to proving subcriticality of disagreement clusters under the grand coupling.

Abstract: Denoising diffusion probabilistic models (DDPMs) have emerged as a powerful framework for generative modeling by transforming standard Gaussian inputs into outputs that approximate a target distribution. While DDPMs have been proved to converge in total variation, KL divergence, and Wasserstein distance, these discrepancy measures share a fundamental limitation of being insensitive to distributional mismatch in low-probability regions. To address this, we adopt the stronger Rényi divergence as the discrepancy measure. Leveraging a Girsanov-type change-of-measure argument, we establish the first Rényi characterization of DDPMs to the best of our knowledge, showing that the total error decomposes into a mixing error, a discretization error, and a score estimation error. We assume the target distribution is sub-Gaussian and characterize the score error via the cumulant generating function, relaxing both the log-Sobolev-inequality and $L^\infty$-score-error conditions used in prior Rényi divergence analyses of inexact Langevin Monte Carlo, a score-based sampling algorithm closely related to DDPM. Our results also establish an iteration complexity for achieving $ε$-accuracy in $α$-Rényi divergence that is asymptotically proportional to $α^3/ε$. Finally, by extending our analysis, we establish the first Rényi divergence convergence guarantee for critically damped DDPMs, a kinetic variant governed by critically damped Langevin dynamics. Similar to existing characterizations of critically damped DDPMs, this result shares the structural scaling of standard DDPMs, differing by at most a logarithmic factor in the mixing error exponent, and supports the conjecture that critically damped DDPMs do not yield a theoretical improvement.

Abstract: Sampling temperature can substantially reshape autoregressive generation, but it is unclear when temperature changes in long-range correlations reflect a genuine phase transition rather than a smooth change in statistics. We study this question in a controlled analogue of temperature sampling: a finite-depth probabilistic context-free grammar on a fixed hierarchical tree, where exact next-token conditionals are computed by belief propagation and then rescaled by a softmax temperature before autoregressive sampling. We measure two-point mutual information between tokens as a function of separation. Across the tested temperatures and grammar ensembles, mutual information remains well described by a power-law envelope. Temperature changes the correlation magnitude and fitted exponent, but within the accessible system sizes we find no long-distance plateau characteristic of an ordered phase and no clear crossover to short-range exponential decay. An approximate mapping to an Ising model on a tree explains how exponential decay in tree distance becomes a power law in token distance, allowing power laws to persist without tuning to a special decoding temperature. This controlled baseline clarifies which temperature effects can arise from hierarchical structure alone, before adding training, approximation error, or neural architecture.

Abstract: We show that the 2D classical spin model constructed in arXiv:2605.10943 hosts many metastable states at non-zero temperature which are stable to arbitrary Hamiltonian perturbations, providing the first such example. Our proof relies on a recursive application of the cluster expansion.

Abstract: We show that the guaranties of the Gram-Schmidt Walk Algorithm of Bansal, Dadush, Garg, and Lovett can be achieved by an Ising model. Given vectors $v_1, \ldots, v_n$ of norm less than 1, we give a closed form for a centrally symmetric distribution on a vector $σ\in \{\pm 1 \}^n$ so that the signed sum $\sum_j σ(j) v_j$ is $1$-subgaussian. We show that the Glauber dynamics for this model mix rapidly provided that the norms of the input vectors are bounded below 1 by an inverse-polynomial. The proofs employ the theory of Lorentzian polynomials, and we use these techniques to show fast mixing bounds for other spin systems.

Abstract: For $Δ\ge2$ and $q\ge11Δ/6$, we prove that the antiferromagnetic $q$-state Potts partition function on finite simple graphs of maximum degree at most $Δ$ has no Fisher zeros in a graph-uniform complex neighbourhood of $[0,1]$. For $q>11Δ/6$, the proof establishes coupling independence throughout $[0,1]$ using a soft version of Vigoda's flip dynamics. Our main tool is a separator-shell transfer theorem for the Potts model on induced-subgraph closed classes of graphs of maximum degree at most $Δ$, with $q\geΔ+1$. It yields a graph-uniform zero-free neighbourhood of $[0,1]$ from Hamming coupling independence at $0$ and a uniform coupling-independence bound on each interval $[δ,1]$, $δ\in(0,1]$. We also obtain zero-free Lee-Yang polydiscs around the uniform field for vertex- and edge-colour fields. For Boolean Holant problems on bounded-degree graphs whose signatures come from a fixed finite family of log-concave symmetric signatures $f$ with $f(0)>0$, such as $b$-matchings, we obtain graph-uniform zero-free polytubes around every bounded box of nonnegative activities; their union is an open zero-free neighbourhood of the nonnegative orthant. An appendix summarizes further coupling-independence inputs and the zero-free regions they yield.

Abstract: We give a polynomial-time algorithm for approximating the partition function of mean-field mixed $p$-spin models to arbitrarily high accuracy throughout the second-moment regime. This improves the quasipolynomial-time algorithm of Bencs, Huang, Lee, Liu, and Regts (arXiv:2507.15616) and answers their open question. In particular, our result covers the entire replica-symmetric regime of the Sherrington--Kirkpatrick model. The algorithm is deterministic, runs in time polynomial in $n$ and $1/\varepsilon$, and succeeds for every typical realization of the disorder. Our main technical contribution is a new algorithmic application of color coding to the classical high-temperature expansion introduced by Aizenman, Lebowitz, and Ruelle in their study of fluctuations of the Sherrington--Kirkpatrick partition function. We combine this approach with the zero-freeness established by Bencs et al. (arXiv:2507.15616) to convert additive approximations into multiplicative ones.

Abstract: We design the first polynomial-time algorithms for approximately counting and almost uniformly sampling common bases of two matroids given by their independence oracles. Moreover, our algorithms generalize far beyond this to Hadamard products of two probability measures on the Boolean cube satisfying a simple nonnegative curvature condition. These algorithmic primitives have myriad applications in statistical physics, polyhedral combinatorics, the study of quantum many-body systems, and beyond. Our approach has two key ingredients. $\bullet$ We relax the intersection by imposing an overlap penalty on the product measure formed by the two input measures. We prove, via an integrated Bochner-type method, that this "$\textit{soft intersection}$" satisfies a Poincare inequality uniformly over all external fields. $\bullet$ We solve a dual maximum entropy convex program to compute external fields under which the hard constraint is satisfied with high probability under the soft intersection measure. We bound this success probability directly using the uniform Poincare inequality and smallness of the gradient norm. $\textbf{AI Disclosure}$ GPT-5.6 Sol Ultra and GPT-6 Astra Ultra were heavily used to develop the ideas in this paper. A more complete discussion is included in the acknowledgments.

Abstract: Diffusion models have emerged as state-of-the-art generative models, with recent extensions from Euclidean spaces to Riemannian manifolds. However, existing convergence guarantees for Riemannian diffusion models typically require $\tilde{O}(\mathrm{poly}(d,T)/ε^2)$ score evaluations, with potentially unfavorable dependence on the dimension. In this work, we develop a general framework that separates score discretization from Brownian-motion simulation and allows multiple geodesic random-walk steps per score evaluation. Under nonnegative Ricci curvature assumption and an exact Brownian-motion simulation oracle, we show that $\tilde{O}(d/ε^2)$ score evaluations suffice to achieve an $ε^2$ KL divergence from the target distribution, matching the existing convergence rate of Euclidean diffusion models. We further show that $\tilde{O}(d^4T/ε^2)$ geodesic random-walk steps suffice to approximate the required drifted Brownian motion to $ε$ total variation error. Combining these results yields a sampling scheme with $\tilde{O}(d/ε^2)$ score evaluations and $\tilde{O}(d^4T/ε^2)$ geodesic random-walk steps, motivating multiple random-walk steps between consecutive score evaluations. Our results provide a sharper characterization of the convergence and sampling complexity of Riemannian diffusion models.

Abstract: Local-to-global techniques for establishing spectral gaps have played a central role in the modern theory of Markov chain mixing times and the theory of high-dimensional expanders. One of the most striking results in this burgeoning literature is that a spectral gap for the global down-up walk on the facets of a pure simplicial complex can be reduced to sufficiently strong spectral expansion of just the codimension-2 links of the complex, a phenomenon colloquially referred to as "trickle-down". These types of theorems have had many important applications, including rapid mixing of the exchange walk on the bases of any matroid. In this primarily expository article, we give streamlined proofs of two such theorems in the literature, one by Oppenheim (2018) and one by Leake and Oveis Gharan (2025), via an integrated Bochner method. Moreover, in the latter setting, we quantitatively strengthen the dependence of the global spectral gap on the dimension of the complex and the spectral influence, resolving an open question of Leake and Oveis Gharan. Disclaimer: The proofs were developed through a couple of rounds of interaction with GPT-5.6 Sol Ultra. We later discovered that Guo and Zhang (2026) had independently proven the same strengthening of the trickle-down theorem of Leake and Oveis Gharan using an extremely similar argument, also found by GPT-5.6 Sol Ultra. The focus of their paper is the complexity of approximating the partition function of spin systems on planar graphs, not on the trickle-down phenomenon itself. In contrast, our motivation is primarily expository, and we hope to bring Bochner-type methods and their connections with the trickle-down phenomenon to the attention of a wider community of researchers.

Abstract: We establish a degree-independent bound on spectral independence for log-concave Holant problems on simple graphs. As a corollary, we obtain relaxation-time bounds for Glauber dynamics of $O_λ(m)$ for the monomer-dimer model at activity $λ$ and $O_{b,λ}(m)$ for $b$-matchings at fugacity $λ>0$, where $m$ is the number of edges. For uniform $b$-matchings, the relaxation-time bound improves to $O(bm)$. The main proof ideas were found using GPT-5.6 Sol.

Abstract: Sparsity is a powerful structural resource in optimization and statistics. We develop frameworks for leveraging sparsity in sampling problems over the Hamming slice $\mathcal{X}_k^d:=\{\mathbf{x}\in\{\pm 1\}^d:|\{i:\mathbf{x}_i=1\}|=k\}$, in high-dimensional regimes where $k\ll d$ (i.e., where $\mathcal{X}_k^d$ is \emph{highly magnetized}). We use our frameworks to design improved samplers for canonical problems in the study of \emph{Ising models} and \emph{Bayesian sparse linear regression}. Our first main result considers the \emph{Sherrington--Kirkpatrick} (SK) model restricted to fixed-magnetization slices $\mathcal{X}_k^d$. We give a polynomial-time sampler for fixed-magnetization SK models at any inverse temperature $β>0$, under arbitrary external fields, provided that $k\le c_βd$ for an appropriate constant $c_β$. By combining this result with an annealing strategy for estimating normalizing constants, we obtain polynomial-time samplers for the SK model at arbitrarily low temperatures under a sufficiently strong external field of strength $h$. In the large-$β$ limit, our framework permits sampling at field strengths within constant factors of the \emph{Almeida--Thouless line} delineating the replica-symmetric and replica-symmetry-breaking regions ([dAT78]), improving polynomially over the field strength $h(β)$ required by the recent work of [BAR26]. Our second main result concerns the measurement complexity of polynomial-time Bayesian sparse linear regression. Recent work by [KSTZ25] shows how to sample from the canonical \emph{Gaussian spike-and-slab posterior} with expected sparsity $k$, at any signal-to-noise ratio, given $n\gtrsim k^3\log^3 d$ Gaussian measurements. We improve this requirement to $n\gtrsim k^{3/2}\log^2 d+k\log^3 d$, using a common sparsity-aware framework underlying both our results.