DMRG Preprints

While epidemiological modeling is pivotal for informing public health strategies, a fundamental trade-off limits its predictive fidelity: exact stochastic simulations are often computationally intractable for large-scale systems, whereas efficient analytical approximations typically fail to account for essential short-range correlations and network loops. Here, we resolve this trade-off by introducing Tensor Network Dynamical Message Passing (TNDMP), a framework grounded in a rigorous property we term \textit{Susceptible-Induced Factorization}. This theoretical insight reveals that a susceptible node acts as a dynamical decoupler, factorizing the global evolution operator into localized components. Leveraging this, TNDMP provides a dual-mode algorithmic suite: an exact algorithm that computes local observables with minimal redundancy on tractable topologies and a scalable and tunable approximation for complex real-world networks. We demonstrate that widely adopted heuristics, such as Dynamical Message Passing (DMP) and Pair Approximation (PA), are mathematically recoverable as low-order limits of our framework. Numerical validation in synthetic and real-world networks confirms that TNDMP significantly outperforms existing methods to predict epidemic thresholds and steady states, offering a rigorous bridge between the efficiency of message passing and the accuracy of tensor network formalisms.

We study the elementary excitations and magnetization of a one-dimensional spin-1/2 XX chain comprising trimer-dimer units (the J1-J1-J2-J3-J2 topology) under a transverse magnetic field h. Using Green's function theory and the Jordan-Wigner transformation, we map the system onto spinless fermions and focus on antiferromagnetic (AFM) interactions. At zero temperature, distinct 1/5 and 3/5 magnetization plateaus emerge, determined by the global periodicity Q=5, with the number of plateaus matching the number of excitation gaps above the Fermi level of the spinless fermions. The magnetic phase diagram in the (h-Js) plane features a Luttinger liquid (LL) state, a gapless AFM state, two magnetization plateau states, and a fully polarized gapped magnetic state. The widths of the LL and gapless AFM phases are found to be proportional to the bandwidths gamma = |E(k=0)-E(k=pi)| of the corresponding elementary excitations, whereas the widths of the magnetization plateau states are governed by the excitation gaps. Our study opens new directions for exploring interacting trimer-dimer spin chains in quantum magnetism using experimental techniques such as neutron scattering, as well as theoretical and numerical approaches including quantum Monte Carlo (QMC) and density-matrix renormalization group (DMRG) methods. Furthermore, we extend the Oshikawa-Yamanaka-Affleck (OYA) condition to generalized cluster chains, demonstrating that the allowed magnetization plateaus are governed by the global periodicity of the chain (e.g., Q=5 for a trimer-dimer chain), rather than by the local periodicity of individual units (Q=3 for a trimer or Q=2 for a dimer).

Published in Physical Review B 113, 064401 (2026)

We demonstrate that the infinite-$U$ triangular-lattice Hubbard model supports a superconducting state built from tightly bound Cooper pairs composed of two holes and one magnon ($2h1m$). Building on the seminal prediction of repulsively bound $2h1m$ states, we show that next-nearest-neighbor hopping $t_{2}$ coherently mixes symmetry-related configurations, stabilizing an $s$-wave bound state with substantial binding energy and a light effective mass. Large-scale DMRG calculations at finite doping identify a magnetization plateau corresponding to a gas of such bound states and quasi--long--range superconducting order with power-law $2h1m$ pair correlations. Our results establish a magnon-mediated superconducting mechanism driven by kinetic frustration, with immediate detectable signatures for moiré Hubbard materials and ultracold-atom simulators.

Machine learning in clinical settings must balance predictive accuracy, interpretability, and privacy. Models such as logistic regression (LR) offer transparency, while neural networks (NNs) provide greater predictive power; yet both remain vulnerable to privacy attacks. We empirically assess these risks by designing attacks that identify which public datasets were used to train a model under varying levels of adversarial access, applying them to LORIS, a publicly available LR model for immunotherapy response prediction, as well as to additional shallow NN models trained for the same task. Our results show that both models leak significant training-set information, with LRs proving particularly vulnerable in white-box scenarios. Moreover, we observe that common practices such as cross-validation in LRs exacerbate these risks. To mitigate these vulnerabilities, we propose a quantum-inspired defense based on tensorizing discretized models into tensor trains (TTs), which fully obfuscates parameters while preserving accuracy, reducing white-box attacks to random guessing and degrading black-box attacks comparably to Differential Privacy. TT models retain LR interpretability and extend it through efficient computation of marginal and conditional distributions, while also enabling this higher level of interpretability for NNs. Our results demonstrate that tensorization is widely applicable and establishes a practical foundation for private, interpretable, and effective clinical prediction.

The realization of fault-tolerant quantum computers remains a challenging endeavor, forcing state-of-the-art quantum hardware to rely heavily on noise mitigation techniques. Standard quantum error mitigation is typically based on post-processing strategies. In contrast, the present work explores a pre-processing approach, in which the effects of noise are mitigated before performing a measurement on the output state. The main idea is to find an observable $Y$ such that its expectation value on a noisy quantum state $\mathcal{E(ρ)}$ matches the expectation value of a target observable $X$ on the noiseless quantum state $ρ$. Our method requires the execution of a noisy quantum circuit, followed by the measurement of the surrogate observable $Y$. The main enablers of our method in practical scenarios are Tensor Networks. The proposed method improves over Tensor Error Mitigation (TEM) in terms of average error, circuit depth, and complexity, attaining a measurement overhead that approaches the theoretical lower bound. The improvement in terms of classical computation complexity is in the order of $\sim 10^6$ times when compared to the post-processing computational cost of TEM in practical scenarios. Such gain comes from eliminating the need to perform the set of informationally complete positive operator-valued measurements (IC-POVM) required by TEM, as well as any other tomographic strategy.

Frustration on the triangular lattice has long been a source of intriguing and often debated phases in many-body systems. Although symmetry analysis has been employed, the role of the seemingly trivial parity symmetry has received little attention. In this work, we show that phases induced by frustration are systematically shaped by an implicit rule of thumb associated with spontaneous parity breaking. This principle enables us to anticipate and rationalize the regimes and conditions under which nontrivial phases emerge. For the spin-$S$ antiferromagnetic XXZ model, we demonstrate that a controversial parity-broken phase appears only at intermediate values of $S$. In bilayer systems, enhanced frustration leads to additional phases, such as supersolids, whose properties can be classified by their characteristic parity features. Benefiting from our improved tensor network contraction techniques, we confirm these results through large-scale tensor-network calculations. This study offers an alternative viewpoint and a systematic approach for examining the interplay between spin, symmetry, and frustration in many-body systems.

Improving the computational efficiency of quantum many-body calculations from a hardware perspective remains a critical challenge. Although field-programmable gate arrays (FPGAs) have recently been exploited to improve the computational scaling of algorithms such as Monte Carlo methods, their application to tensor network algorithms is still at an early stage. In this work, we propose a fine-grained parallel tensor network design based on FPGAs to substantially enhance the computational efficiency of two representative tensor network algorithms: the infinite time-evolving block decimation (iTEBD) and the higher-order tensor renormalization group (HOTRG). By employing a quad-tile partitioning strategy to decompose tensor elements and map them onto hardware circuits, our approach effectively translates algorithmic computational complexity into scalable hardware resource utilization, enabling an extremely high degree of parallelism on FPGAs. Compared with conventional CPU-based implementations, our scheme exhibits superior scalability in computation time, reducing the bond-dimension scaling of the computational cost from $O(D_b^3)$ to $O(D_b)$ for iTEBD and from $O(D_b^6)$ to $O(D_b^2)$ for HOTRG. This work provides a theoretical foundation for future hardware implementations of large-scale tensor network computations.

This report documents the program of the second Toulouse Tensor Workshop which took place at the University of Toulouse on September 17-19, 2025, and summarizes the main points of discussion. This workshop follows the first Workshop (CECAM workshop on Tensor Contraction Library Standardization), which took place in Toulouse one year earlier, on May 24-25, 2024 and led to the formation of a tensor standardization working group, which has since specified a low-level standard interface for tensor operations available freely on GitHub. The 2025 workshop brought together developers of applications which rely extensively on tensor computations such as quantum many-body simulations in chemistry and physics (material science and electronic structure calculations), as well as developers and experts of tensor software who have the know-how to provide the technical support for such applications. The workshop enabled the community to provide feedback on the specified low-level interface and how it can be further refined. It also initiated a discussion on how the standardization efforts should be oriented in the near feature, in particular on what should be higher-level interfaces and how to tackle other requirements of the community such as tensor decompositions, symmetric tensors and structured sparsity support.

Characterizing the ground-state properties of disordered systems, such as spin glasses and combinatorial optimization problems, is fundamental to science and engineering. However, computing exact ground states and counting their degeneracies are generally NP-hard and #P-hard problems, respectively, posing a formidable challenge for exact algorithms. Recently, Tensor Networks methods, which utilize high-dimensional linear algebra and achieve massive hardware parallelization, have emerged as a rapidly developing paradigm for efficiently solving these tasks. Despite their success, these methods are fundamentally constrained by the exponential growth of space complexity, which severely limits their scalability. To address this bottleneck, we introduce the Branch-and-Bound Tensor Network (BBTN) method, which seamlessly integrates the adaptive search framework of branch-and-bound with the efficient contraction of tropical tensor networks, significantly extending the reach of exact algorithms. We show that BBTN significantly surpasses existing state-of-the-art solvers, setting new benchmarks for exact computation. It pushes the boundaries of tractability to previously unreachable scales, enabling exact ground-state counting for $\pm J$ spin glasses up to $64 \times 64$ and solving Maximum Independent Set problems on King's subgraphs up to $100 \times 100$. For hard instances, BBTN dramatically reduces the computational cost of standard Tropical Tensor Networks, compressing years of runtime into minutes. Furthermore, it outperforms leading integer-programming solvers by over 30$\times$, establishing a versatile and scalable framework for solving hard problems in statistical physics and combinatorial optimization.