arXiv subjects

Sumit Mukherjee

Publications and source records attributed to Sumit Mukherjee.

At least 19 recordsLinked to original sources

arXiv2026

LDP for Tensor Forms

In this paper, we study the large deviation principle (LDP) for a tensor-weighted functional of i.i.d. random variables, when the sequence of tensors converges under a variant of the "bad" cut norm. Using the LDP, we analyze a Gibbs measure with a tensor-valued Hamiltonian, and characterize the optimizers of the limiting variational problem in terms of a functional fixed point equation. As applications, we focus on several concrete examples, which include monochromatic subgraph counts in sparse random graphs, Erd\H{o}s-R\'enyi hypergraphs, and a generalized Potts statistic of order $v\ge 2$. Studying the optimization problem, we give sufficient conditions for uniqueness of the optimizer, as well as for existence of constant optimizers (replica symmetry). Our results demonstrate universal weak laws for a large class of tensor Gibbs models with approximately regular tensors.

math.PR↗

arXiv2026

Joint parameters estimation in cubic tensor model

We study joint parameter estimation from a single observation in high-dimensional Gibbs measures with cubic tensor interactions, motivated by dense ERGMs, arithmetic-progression models, and inhomogeneous random hypergraphs. Focusing on the maximum pseudolikelihood estimator, we give checkable conditions for joint consistency and asymptotic ill-conditioning. For the edge-triangle ERGM, pseudolikelihood is ill-conditioned in the ferromagnetic regime with nonnegative field, but consistent in a sufficiently strong antiferromagnetic regime. For the edge-three-star ERGM, it is ill-conditioned for all inverse temperatures and external fields. We also study consistency for arithmetic-progression, and inhomogeneous hypergraph models. Our proofs develop nonlinear large-deviation and mean-field approximation tools for cubic tensor Gibbs measures, which have scope for broad applications.

math.ST↗

arXiv2026

RASC+: Retrieval-Constrained LLM Adjudication for Clinical Value Set Authoring

Clinical value sets define the standardized terminology codes used in quality measurement, phenotyping, cohort construction, and clinical decision support. The recently introduced Retrieval-Augmented Set Completion (RASC) benchmark showed that direct zero-shot large language model (LLM) generation is poorly suited to this task: clinical code systems are large, version-controlled, and not reliably memorized by language models. We study a stage-wise alternative in which candidate-pool construction is optimized for recall and a constrained LLM adjudicator is optimized for candidate selection. On the full 3,744-value-set RASC test split, Qwen3-based retrieval with vocabulary-aware expansion and code-display rescue retrieval increases candidate-pool recall from the original RASC retrieval baseline of 0.553 to 0.730; on the held-out-publisher stratum, pool recall is 0.655. The higher-recall pool alone is not sufficient: applying the original SAPBert cross-encoder to this expanded pool gives full-test macro F1 of 0.287 and held-out-publisher macro F1 of 0.233. Replacing the stage-2 selector with blinded GPT-5 adjudication over the same pool increases full-test macro F1 to 0.549 and held-out-publisher macro F1 to 0.533. These results show that retrieval-constrained LLM adjudication can substantially improve value set completion while preserving the safety constraint that all returned codes must come from an auditable candidate pool.

cs.CL↗

arXiv2026

Intraparticle entanglement-based Random Access Code protocols: Contextuality-enabled quantum advantage and implications

We provide the first explicit identification and quantitative characterization of the physical origin of the quantum advantage in the Random Access Code (RAC) protocol. This is achieved by formulating the protocol in terms of intraparticle entanglement between co-measurable degrees of freedom of a single particle and establishing a fundamental correspondence between the protocol's success probability and the underlying resource powering it. For this purpose, we use a relevant Bell-type inequality derived from the assumption of noncontextuality of measurement outcomes. The formulated analysis reveals that the magnitude of quantum-mechanical violation of this inequality, signifying a form of quantum contextuality, is quantitatively commensurate with the ``quantum enhancement" of success probability in any intraparticle entanglement-assisted $n$-bit RAC protocol. In particular, the maximal success probability achievable in a quantum $n \mapsto 1$ RAC protocol corresponds to the maximal quantum violation of the relevant Bell-type inequality. Our framework not only demonstrates how quantum contextuality entailed by intraparticle entanglement serves as an effective resource for enhancing RAC performance, but also offers a significant operational advantage: the proposed scheme is readily implementable in a single-particle interferometric setup requiring coherence preservation only for a single particle, rather than between spatially separated entangled systems.

quant-ph↗

arXiv2026

Geometric Comparisons of Electoral Rules Under Feedback

We study how electoral rules shape polarization dynamics when voters and candidates both adapt to repeated election outcomes. We introduce two geometric primitives for comparing rules under this feedback: the \emph{winner radius} $R_t = \max_i \|x_i - w^{(t)}\|$, the distance from the winner to the farthest voter, and the \emph{supporter centroid radius} $S_t = \max_j \|c_j - s_j^{(t)}\|$, the largest gap between any candidate and their support base. We show that $R_t$ controls a one-step contraction bound on voter disagreement and $S_t$ plays the analogous role for candidate dispersion, and that these two objectives are in tension. Rules that reduce $R_t$ tend to increase $S_t$, and vice versa. A winner close to the voter median does not resolve the tension, since proximity to the median and proximity to the Chebyshev center are different objectives. We use this framing to organize a simulation study across seven standard electoral rules and one convex-combination benchmark, comprising 1000+ runs across diverse electorate profiles, voter mechanisms, and camp-balance settings. The empirical results confirm the theoretical tradeoff: winner-take-all rules achieve small $S_t$ at the cost of large $R_t$ and weaker voter depolarization, while convex-combination rules reverse this. An oracle comparison further shows that minimizing $R_t$ per step and minimizing voter disagreement per step are distinct objectives with different long-run consequences for both voter and candidate dynamics.

cs.GT↗

arXiv2026

Retrieve, Then Classify: Corpus-Grounded Automation of Clinical Value Set Authoring

Clinical value set authoring -- the task of identifying all codes in a standardized vocabulary that define a clinical concept -- is a recurring bottleneck in clinical quality measurement and phenotyping. A natural approach is to prompt a large language model (LLM) to generate the required codes directly, but structured clinical vocabularies are large, version-controlled, and not reliably memorized during pretraining. We propose Retrieval-Augmented Set Completion (RASC): retrieve the $K$ most similar existing value sets from a curated corpus to form a candidate pool, then apply a classifier to each candidate code. Theoretically, retrieve-and-select can reduce statistical complexity by shrinking the effective output space from the full vocabulary to a much smaller retrieved candidate pool. We demonstrate the utility of RASC on 11,803 publicly available VSAC value sets, constructing the first large-scale benchmark for this task. A cross-encoder fine-tuned on SAPBert achieves AUROC~0.852 and value-set-level F1~0.298, outperforming a simpler three-layer Multilayer Perceptron (AUROC~0.799, F1~0.250) and both reduce the number of irrelevant candidates per true positive from 12.3 (retrieval-only) to approximately 3.2 and 4.4 respectively. Zero-shot GPT-4o achieves value-set-level F1~0.105, with 48.6\% of returned codes absent from VSAC entirely. This performance gap widens with increasing value set size, consistent with RASC's theoretical advantage. We observe similar performance gains across two other classifier model types, namely a cross-encoder initialized from pre-trained SAPBert and a LightGBM model, demonstrating that RASC's benefits extend beyond a single model class. The code to download and create the benchmark dataset, as well as the model training code is available at: \href{https://github.com/mukhes3/RASC}{https://github.com/mukhes3/RASC}.

cs.CL↗

arXiv2026

Joint Estimation in Potts Model

In this paper, we study estimation of parameters in a two-parameter Potts model with $q$ colors and coupling matrix $A_N$. We characterize concrete sufficient conditions for existence of the pseudo-likelihood estimator of the Potts model, in terms of the local magnetic fields, and give sufficient conditions for the validity of the above characterization. We then provide sufficient criteria for estimation of both parameters at the optimal rate $\sqrt{N}$. In particular, if $A_N$ is the scaled adjacency matrix of a graph $G_N$, then we show that joint estimation is possible if either $G_N$ has bounded degree or is irregular. In contrast, we give an example of a graph sequence $G_N$ which is approximately regular and dense, where no consistent estimator exists. We also show that one-parameter estimation at the optimal rate $\sqrt{N}$ holds under much milder conditions when the other parameter is known. Along the way, we develop a concentration result for mean-field Potts models using the framework of nonlinear large deviations. Compared to the Ising case, our results for the Potts case require a novel analysis across multiple colors.

math.ST↗

arXiv2026

Connection between the contextuality breaking and incompatibility breaking qubit channels

Contextuality and measurement incompatibility are two fundamental aspects of nonclassicality, and their manifestations in observed quantum correlations are often deeply interconnected. Recently, measurement incompatibility has been studied in connection with nonlocality, particularly in terms of their robustness under various quantum channels. This line of investigation helps establish a connection between the channels that break nonlocality and those that break incompatibility. In this study, we focus on an asymmetric bipartite Bell scenario involving three and four inputs on Alice and Bob sides, respectively, with each of these inputs having dichotomous outcomes. Under the assumption of locality, the observed statistics in this asymmetric scenario obeys the Elegant Bell inequality (EBI). Here, we use a different version of the EBI that relies on the assumption of the preparation noncontextuality. By taking the violation of this noncontextual version of EBI as a witness of preparation contextuality we establish a connection between the channels that break contextuality and the channels that break triple-wise measurement incompatibility. Our results suggest that any channel which breaks EBI contextuality will also break Clauser-Horne-Shimony-Holt (CHSH) nonlocality; however, the reverse does not hold. We also show that a depolarising channel that breaks N-wise incompatibility can also break a certain form of contextuality, witnessed by a generalised inequality involving N measurements on one wing of a bipartite Bell scenario.

quant-ph↗

arXiv2026

Electoral Systems Simulator: An Open Framework for Comparing Electoral Mechanisms Across Voter Distribution Scenarios

Here we present \texttt{electoral\_sim}, an open-source Python framework for simulating and comparing electoral systems across diverse voter preference distributions. The framework represents voters and candidates as points in a two-dimensional ideological space, derives sincere ballot profiles from Euclidean preference distances, and evaluates several standard electoral mechanisms -- including plurality, ranked-choice, approval, score, Condorcet, and two proportional representation systems -- against a common primary metric: the Euclidean distance between the electoral outcome and the geometric median of the voter distribution. We evaluate these systems across many empirically-grounded scenarios ranging from unimodal consensus electorates to sharply polarised bimodal configurations, reporting both single-run and Monte Carlo stability results across 200 trials per scenario. As a case study in framework extensibility, we implement and evaluate a novel hypothetical mechanism that is not currently implemented in any jurisdiction -- in which each voter's influence is distributed across candidates via a Boltzmann softmax kernel. This system is included as a theoretical benchmark characterising an approximate upper bound on centroid-seeking performance, rather than as a policy proposal. All code is released publicly at https://github.com/mukhes3/electoral_sim.

cs.GT↗

arXiv2025

Limits of Absoluteness of Observed Events in Timelike Scenarios: A No-Go Theorem

Wigner's Friend-type paradoxes challenge the assumption that events are absolute -- that when we measure a system, we obtain a single result, which is not relative to anything or anyone else. These paradoxes highlight the tension between quantum theory and our intuitions about reality being observer-independent. Building on a recent result that developed these paradoxes into a no-go theorem, namely the Local Friendliness Theorem, we introduce the Causal Friendliness Paradox, a time-ordered analogue of it. In this framework, we replace the usual locality assumption with Axiological Time Symmetry (ATS), and show that, when combined with the assumptions of Absoluteness of Observed Events (AOE), No Retrocausality (NRC), and Screening via Pseudo Events (SPE), we obtain a causal inequality. We then show that quantum mechanics violates this inequality and is therefore incompatible with at least one of these assumptions. To probe which assumption might be incompatible, we then examine whether AOE in its entirety is essential for this no-go result. We propose a weaker, operational form of AOE that still leads to inequalities that quantum mechanics violates. This result shows that even under relaxed assumptions, quantum theory resists reconciliation with classical notions of absolute events, reinforcing the foundational significance of Wigner's Friend-type paradoxes in timelike scenarios.

quant-ph↗

arXiv2025

Inference on Gaussian mixture models with dependent labels

Gaussian mixture models are widely used to model data generated from multiple latent sources. Despite its popularity, most theoretical research assumes that the labels are either independent and identically distributed, or follows a Markov chain. It remains unclear how the fundamental limits of estimation change under more complex dependence. In this paper, we address this question for the spherical two-component Gaussian mixture model. We first show that for labels with an arbitrary dependence, a naive estimator based on the misspecified likelihood is $\sqrt{n}$-consistent. Additionally, under labels that follow an Ising model, we establish the information theoretic limitations for estimation, and discover an interesting phase transition as dependence becomes stronger. When the dependence is smaller than a threshold, the optimal estimator and its limiting variance exactly matches the independent case, for a wide class of Ising models. On the other hand, under stronger dependence, estimation becomes easier and the naive estimator is no longer optimal. Hence, we propose an alternative estimator based on the variational approximation of the likelihood, and argue its optimality under a specific Ising model.

math.ST↗

arXiv2025

CLT in high-dimensional Bayesian linear regression with low SNR

We study central limit theorems for linear statistics in high-dimensional Bayesian linear regression with product priors. Unlike the existing literature where the focus is on posterior contraction, we work under a non-contracting regime where neither the likelihood nor the prior dominates the other. This is motivated by modern high-dimensional datasets characterized by a bounded signal-to-noise ratio. This work takes a first step towards understanding limit distributions for one-dimensional projections of the posterior, as well as the posterior mean, in such regimes. Analogous to contractive settings, the resulting limiting distributions are Gaussian, but they heavily depend on the chosen prior and center around the Mean-Field approximation of the posterior. We study two concrete models of interest to illustrate this phenomenon -- the white noise design, and the (misspecified) Bayesian model. As an application, we construct credible intervals and compute their coverage probability under any misspecified prior. Our proofs rely on a combination of recent developments in Berry-Esseen type bounds for Random Field Ising models and both first and second order Poincar\'{e} inequalities. Notably, our results do not require any sparsity assumptions on the prior.

math.ST↗

arXiv2025

Variational Inference for Latent Variable Models in High Dimensions

Variational inference (VI) is a popular method for approximating intractable posterior distributions in Bayesian inference and probabilistic machine learning. In this paper, we introduce a general framework for quantifying the statistical accuracy of mean-field variational inference (MFVI) for posterior approximation in Bayesian latent variable models with categorical local latent variables (and arbitrary global latent variables). Utilizing our general framework, we capture the exact regime where MFVI 'works' for the celebrated latent Dirichlet allocation model. Focusing on the mixed membership stochastic blockmodel, we show that the vanilla fully factorized MFVI, often used in the literature, is suboptimal. We propose a partially grouped VI algorithm for this model and show that it works, and derive its exact finite-sample performance. We further illustrate that our bounds are tight for both the above models. Our proof techniques, which extend the framework of nonlinear large deviations, open the door for the analysis of MFVI in other latent variable models.

math.ST↗

arXiv2025

Distance-based measures and Epsilon-measures for measurement-based quantum resources

Quantum resource theories provide a structured and elegant framework for quantifying quantum resources. While state-based resource theories have been extensively studied, their measurement-based resource theories remain relatively underexplored. In practical scenarios where a quantum state or a set of measurements is only partially known, conventional resource measures often fall short in capturing the resource content. In such cases, \epsilon-measures offer a robust alternative, making them particularly valuable. In this work, we investigate the quantification of measurement-based resources using distance-based measures, followed by a detailed analysis of the mathematical properties of \epsilon-measures. We also extend our analysis by exploring the connections between \epsilon-measures and some key quantities relevant to resource manipulation tasks. Importantly, the analysis of resources based on sets of measurements are tedious compared to that of single measurements as the former allows more general transformations such as controlled implementation. Yet our framework applies not only to resources associated with individual measurements but also to those arising from sets of measurements. In short, our analysis is applicable to existing resource theories of measurements and has the potential to be useful for all resource theories of measurements that are yet to be developed.

quant-ph↗

arXiv2025

Fluctuations in random field Ising models

This paper establishes a CLT for linear statistics of the form $\langle \mathbf{q},\boldsymbol{\sigma} \rangle$ with quantitative Berry-Esseen bounds, where $\boldsymbol{\sigma}$ is an observation from an exponential family with a quadratic form as its sufficient statistic, in the \enquote{high-temperature} regime. We apply our general result to random field Ising models with both discrete and continuous spins. To demonstrate the generality of our techniques, we apply our results to derive both quenched and annealed CLTs in various examples, which include Ising models on some graph ensembles of common interest (Erd\H{o}s-R\'{e}nyi, regular, dense bipartite), and the Hopfield spin glass model. Our proofs rely on a combination of Stein's method of exchangeable pairs and Chevet type concentration inequalities.

math.PR↗

arXiv2025

Contextuality sans incompatibility in the simplest scenario: Communication supremacy of a qubit

Conventional wisdom asserts that measurement incompatibility is necessary for revealing the non-locality and contextuality. In contrast, a recent work [Phys. Rev. Lett. 130, 230201 (2023)] demonstrates the generalized contextuality without measurement incompatibility by using a five-outcome qubit measurement. In this paper, we introduce a two-party prepare-measure communication game involving specific constraints on preparations, and we demonstrate contextuality sans incompatibility in the simplest measurement scenario, requiring only a three-outcome extremal qubit measurement. This contrasts with the aforementioned five-outcome qubit measurement, which can be simulated by an appropriate convex mixture of five three-outcome incompatible qubit measurements. Furthermore, we illustrate that our result has a prominent implication in information theory. Our communication game can be perceived as a constrained Holevo-Frankle-Weiner (HFW) scenario, as operational restrictions are imposed on preparations. We show that the maximum success probability of the game by using a qubit surpasses that attainable by a c-bit, even when shared randomness is a free resource. Consequently, this finding exemplifies the supremacy of a qubit over a c-bit within a constrained HFW framework. Thus, alongside offering fresh insights into quantum foundations, our results pave a novel pathway for exploring the efficacy of a qubit in information processing tasks.

quant-ph↗

arXiv2024

Universality of Persistence of Random Polynomials

We investigate the probability that a random polynomial with independent, mean-zero and finite variance coefficients has no real zeros. Specifically, we consider a random polynomial of degree $2n$ with coefficients given by an i.i.d. sequence of mean-zero, variance-1 random variables, multiplied by an $\frac{\alpha}{2}$-regularly varying sequence for $\alpha>-1$. We show that the probability of no real zeros is asymptotically $n^{-2(b_{\alpha}+b_0)}$, where $b_{\alpha}$ is the persistence exponents of a mean-zero, one-dimensional stationary Gaussian processes with covariance function as $\mathrm{sech}((t-s)/2)^{\alpha+1}$. Our work generalizes the previous results of Dembo et al. [DPSZ02] and Dembo \& Mukherjee [DM15] by removing the requirement of finite moments of all order or Gaussianity. In particular, in the special case $\alpha = 0$, our findings confirm a conjecture by Poonen and Stoll [PS99, Section 9.1] concerning random polynomials with i.i.d. coefficients.

math.PR↗

arXiv2024

Synthesizing Proton-Density Fat Fraction and $R_2^*$ from 2-point Dixon MRI with Generative Machine Learning

Magnetic Resonance Imaging (MRI) is the gold standard for measuring fat and iron content non-invasively in the body via measures known as Proton Density Fat Fraction (PDFF) and $R_2^*$, respectively. However, conventional PDFF and $R_2^*$ quantification methods operate on MR images voxel-wise and require at least three measurements to estimate three quantities: water, fat, and $R_2^*$. Alternatively, the two-point Dixon MRI protocol is widely used and fast because it acquires only two measurements; however, these cannot be used to estimate three quantities voxel-wise. Leveraging the fact that neighboring voxels have similar values, we propose using a generative machine learning approach to learn PDFF and $R_2^*$ from Dixon MRI. We use paired Dixon-IDEAL data from UK Biobank in the liver and a Pix2Pix conditional GAN to demonstrate the first large-scale $R_2^*$ imputation from two-point Dixon MRIs. Using our proposed approach, we synthesize PDFF and $R_2^*$ maps that show significantly greater correlation with ground-truth than conventional voxel-wise baselines.

cs.CV↗