While a previously proposed method for estimating inertial manifold dimension, based on explicitly computing angles between pairs of covariant Lyapunov vectors (CLVs), employs efficient algorithms, it remains computationally demanding due to its substantial resource requirements. In this work, we introduce an improved method to determine this dimension by analyzing the angles between tangent subspaces spanned by the CLVs. This approach builds upon a fast numerical technique for assessing chaotic dynamics hyperbolicity. Crucially, the proposed method requires significantly less computational effort and minimizes memory usage by eliminating the need for explicit CLV computation. We test our method on two canonical systems: the complex Ginzburg-Landau equation and a diffusively coupled chain of Lorenz oscillators. For the former, the results confirm the accuracy of the new approach by matching prior dimension estimates. For the latter, the analysis demonstrates the absence of a low-dimensional inertial manifold, highlighting a complex regime that merits further investigation. The presented method offers a practical and efficient tool for characterizing high-dimensional attractors in extended dynamical systems.

We prove the holding of chaos in the sense of Li-Yorke for a family of four-dimensional discrete dynamical systems that are naturally associated to ODEs systems describing coupled oscillators subject to an external nonconservative force, also giving an example of a discrete map that is Li-Yorke chaotic but not topologically transitive. Analytical results are generalized to a modular definition of the problem and to a system of nonlinear oscillators described by polynomial potentials in one coordinate. We perform numerical simulations looking for a strange attractor of the system; furthermore, we present the bifurcation diagram and perform a bifurcation analysis of the system.

We present analytical and numerical results on integrability and transition to chaotic motion for a generalized Ziegler pendulum, a double pendulum subject to an angular elastic potential and a follower force. Several variants of the original dynamical system, including the presence of gravity and friction, are considered, in order to analyze whether the integrable cases are preserved or not in presence of further external forces, both potential and non-potential. Particular attention is devoted to the presence of dissipative forces, that are analyzed in two different formulations. Furthermore, a study of the discrete version is performed. The analysis of periodic points, that is presented up to period 3, suggests that the discrete map associated to the dynamical system has not dense sets of periodic points, so that the map would not be chaotic in the sense of Devaney for a choice of the parameters that corresponds to a general case of chaotic motion for the original system.

How do the combined effects of phase frustration, noise, and higher-order interactions govern synchronization in globally coupled heterogeneous Kuramoto oscillators? To address this question, we investigate a globally coupled network of Kuramoto-Sakaguchi oscillators that includes both pairwise (1-simplex) and higher-order (2-simplex) interactions, together with additive stochastic forcing. Systematic numerical simulations across a broad range of coupling strengths, phase-lag values, and noise intensities reveal that synchronization emerges through a nontrivial interplay among these parameters. In general, weak frustration combined with mutually reinforcing coupling promotes synchronization, whereas strong frustration favors coherence under repulsive coupling. Forward and backward parameter sweeps reveal the coexistence of synchronized and desynchronized states. The presence and width of this bistable region depend sensitively on phase frustration, noise intensity, and higher-order coupling strength, with higher-order interactions significantly widening the bistable interval. To explain these behaviors, we employ the Ott-Antonsen reduction to derive a low-dimensional amplitude equation that predicts the forward critical point in the thermodynamic limit, the backward saddle-node point, and the width of the bistable region. Higher order interactions widen this region by shifting the saddle-node point without affecting the forward critical point. Further analysis of Kramer's escape rate explains how noise destabilizes coexistence states and diminishes bistability. Overall, our results provide a unified theoretical and numerical framework for frustrated, noisy, higher-order oscillator networks, revealing that synchronization is strongly influenced by the combined action of phase frustration, stochasticity, and both pairwise and higher-order interactions.

The Melnikov method is applied to a class of generalized Ziegler pendulums. We find an analytical form for the separatrix of the system in terms of Jacobian elliptic integrals, holding for a large class of initial conditions and parameters. By working in Duffing approximation, we apply the Melnikov method to the original Ziegler system, showing that the first non-vanishing Melnikov integral appears in the second order. An explicit expression for the Melnikov integral is derived in the presence of a time-periodic external force and for a suitable choice of the parameters, as well as in the presence of a dissipative term acting on the lower rod of the pendulum. These results allow us to define fundamental relationships between the Melnikov integral and a proper control parameter that distinguishes between regular and chaotic orbits for the original dynamical system. Finally, in the appendix, we present proof of a conjecture concerning the non-validity of Devaney's chaoticity definition for a discrete map associated with the system.

In the present research, a bistable delayed-feedback oscillator with two delayed-feedback loops is shown to replicate a network of bistable nodes with nonlocal coupling. It is demonstrated that all the aspects of the nonlocal interaction impact on wavefront propagation identified in networks of bistable elements are entirely reproduced in the dynamics of a single oscillator with two delays. In particular, adding the second delayed-feedback loop allows speeding up both deterministic and stochastic wavefront propagation, achieving stabilization of propagating fronts at lower noise intensity and preventing fronts from noise-induced destruction occurring in the presence of single delayed-feedback. All the revealed effects are studied in numerical simulations and confirmed in physical experiments, showing an excellent correspondence.

Time-delayed phase-oscillator networks model diverse biological and physical systems, yet standard first-order phase reductions cannot adequately capture their high-dimensional collective dynamics. In this Letter, we develop a second-order reduction for a broad class of time-delayed Kuramoto-Daido networks, transforming the original delayed system of one-dimensional phase oscillators into a delay-free network of two-dimensional rotators. The resulting model shows that coupling delay generates inertial terms in the intrinsic dynamics and higher-order (triadic) interactions, and it accurately predicts the emergence of complex collective patterns such as splay, cyclops, and chimera states. The reduction further reveals a qualitative division of roles: time delay acts primarily as effective inertia for higher-dimensional dynamics, including splay states, whereas the induced triadic interactions are decisive for lower-dimensional patterns such as chimeras. The method applies to networks with arbitrary topology, higher-harmonic coupling, and intrinsic-frequency heterogeneity, yielding a compact, parameter-explicit reduced model. This universal reduced description of time-delayed oscillator networks opens the door to systematic prediction and analysis of nontrivial collective dynamics in delay-coupled systems.

Quantum resource theories (QRTs) provide a unified framework for characterizing useful quantum phenomena subject to physical constraints, but are notoriously hard to assess in experimental systems. In this letter, we introduce a unified protocol for quantifying the resource-generating power (RGP) of arbitrary quantum evolutions applicable to multiple QRTs. It is based on deep thermalization (DT), which has recently gained attention for its role in the emergence of quantum state designs from partial projective measurements. Central to our approach is the use of projected ensembles, recently employed to probe DT, together with new twirling identities that allow us to directly infer the RGP of the underlying dynamics. These identities further reveal how resources build up and thermalize in generic quantum circuits. Finally, we show that quantum resources themselves undergo deep thermalization at the subsystem level, offering a complementary and another experimentally accessible route to infer the RGP. Our work connects deep thermalization to resource quantification, offering a new perspective on the essential role of various resources in generating state designs.

Large language model (LLM)-driven agents are emerging as a powerful new paradigm for solving complex problems. Despite the empirical success of these practices, a theoretical framework to understand and unify their macroscopic dynamics remains lacking. This Letter proposes a method based on the least action principle to estimate the underlying generative directionality of LLMs embedded within agents. By experimentally measuring the transition probabilities between LLM-generated states, we statistically discover a detailed balance in LLM-generated transitions, indicating that LLM generation may not be achieved by generally learning rule sets and strategies, but rather by implicitly learning a class of underlying potential functions that may transcend different LLM architectures and prompt templates. To our knowledge, this is the first discovery of a macroscopic physical law in LLM generative dynamics that does not depend on specific model details. This work is an attempt to establish a macroscopic dynamics theory of complex AI systems, aiming to elevate the study of AI agents from a collection of engineering practices to a science built on effective measurements that are predictable and quantifiable.

Vector solitons (VSs), being observed across various fields from optics to Bose-Einstein condensates, are localized structures composed of orthogonal modes bound by nonlinear couplings. Nevertheless, the influence of intermodal linear coupling on the physical properties of this bimodal structure remains to be decently revealed and harnessed. Utilizing an ultrafast fiber laser as a platform, we predict and demonstrate that the linear mode coupling (LMC) induces the deformable VS in terms of the temporal and spectral structures. Weak LMC supports heteronuclear vector solitons built of dissimilar polarization modes, i.e., a single pulse coupled to an orthogonal damped pulse chain. On the other hand, strong LMC facilitates the homonuclear VS composed of polarization modes with similar structures, in the form of soliton compounds featuring caterpillar motions. Our findings reveal new patterns of VSs and open an effective avenue for versatile ultrafast optical sources.

The properties common to the Lorenz and Chen attractors, as well as their fundamental differences, have been studied for many years in a vast number of works and remain a topic far from a rigorous and complete description. In this paper we take a step towards solving this problem by carrying out a rigorous study of the so-called universal normal form to which we have reduced the systems of both of these families. For this normal form, we prove the existence of infinite set of homoclinic orbits with different topological structure defined by the number of rotations around axis of symmetry. We show that these rotational topological features are inherited by the attractors of Chen-type systems and give rise to their twisted nature - the generic difference from attractors of Lorenz type.

We perform an analytical four loop calculation of exponent $z$ in model A of critical dynamics in $d=4-2\varepsilon$ dimensions. This is the first time such a large order of perturbation theory has been calculated analytically for models of critical dynamics. To do this, we apply the modern method of parametrical integration with hyperlogaritms. We discuss in detail peculiarities of application of this method to critical dynamics, e.g. the problem of linear-irreducible diagrams already present in four loop (contrary to statics where the first linear-irreducible diagram appears in six loop).

The inviscid, partial differential equations of hydrodynamics when projected via a Galerkin-truncation on a finite-dimensional subspace spanning wavenumbers $-{\bf K}_{\rm G} \le {\bf k} \le {\bf K}_{\rm G}$, and hence retaining a finite number of modes $N_{\rm G}$, lead to absolute equilibrium states. We review how the Galerkin-truncated, three-dimensional, incompressible Euler equation thermalises and its connection to questions in turbulence. We also discuss an emergent pseudo-dissipation range in the energy spectrum and the time-scales associated with thermalisation.

In dynamical systems on networks, one assigns the dynamics to nodes, which are then coupled via links. This approach does not account for group interactions and dynamics on links and other higher dimensional structures. Higher-order network theory addresses this by considering variables defined on nodes, links, triangles, and higher-order simplices, called topological signals (or cochains). Moreover, topological signals of different dimensions can interact through the Dirac-Bianconi operator, which allows coupling between topological signals defined, for example, on nodes and links. Such interactions can induce various dynamical behaviors, for example, periodic oscillations. The oscillating system consists of topological signals on nodes and links whose dynamics are driven by the Dirac-Bianconi coupling, hence, which we call it Dirac-Bianconi driven oscillator. Using the phase reduction method, we obtain a phase description of this system and apply it to the study of synchronization between two such oscillators. This approach offers a way to analyze oscillatory behaviors in higher-order networks beyond the node-based paradigm, while providing a ductile modeling tool for node- and edge-signals.

Self-organizing systems consume energy to generate internal order. The concept of thermodynamic efficiency, drawing from statistical physics and information theory, has previously been proposed to characterise a change in control parameter by relating the resulting predictability gain to the required amount of work. However, previous studies have taken a system-centric perspective and considered only single control parameters. Here, we generalise thermodynamic efficiency to multiple control parameters and extend the definition of thermodynamic efficiency to protocols in arbitrary directions, by introducing directional efficiency. Taking an observer-centric perspective, we derive two novel formulations. The first, an inferential form, relates efficiency to fluctuations of macroscopic observables, interpreting thermodynamic efficiency in terms of how well the system parameters can be inferred from observable macroscopic behaviour. The second, an information-geometric form, expresses efficiency in terms of the Fisher information matrix, interpreting it with respect to how difficult it is to navigate the statistical manifold defined by the control protocol. This observer-centric perspective is contrasted with the existing system-centric view, where efficiency is considered an intrinsic property of the system.

We review the discrete evolution problem and the corresponding solution as a discrete Dyson series in order to rigorously derive a generalized discrete version of the Magnus expansion. We also systematically derive the discrete analogue of the pre-Lie Magnus expansion and express the elements of the discrete Dyson series in terms of a tridendriform algebra binary operation. In the generic discrete case, extra significant terms that are absent in the continuous or the linear discrete case appear in both Dyson and Magnus expansions. Based on the rigorous discrete derivation key links between quantum algebras, tridendriform and pre-Lie algebras are then established. This is achieved by examining tensor realizations of quantum groups, such as the Yangian. We show that these realizations can be expressed in terms of tridendriform and pre-Lie algebras. The continuous limit as expected provides the corresponding non-local charges of the Yangian as members of the pre-Lie Magnus expansion.

We propose a generalization of the False Vacuum Skyrme model for any simple compact Lie groups $G$ that leads to Hermitian symmetric spaces. The Skyrme field that in the original theory takes its values in $SU(2)$ Lie group, now takes its values in $G$, while the remaining fields correspond to the entries of a symmetric, positive, and invertible $\dim G \times \dim G$-dimensional matrix $h$. This model is also an extension of the generalized BPS Skyrme model. We prove that the global minima correspond to the $h$ fields being self-dual solutions of the generalized BPS Skyrme model, together with a particular field configuration for the Skyrme field that leads to a spherically symmetric topological charge density. As in the case of the original model, the minimization of the energy leads to the so-called reduced problem, defined in the context of false vacuum decay. This imposes a condition on the Skyrme field, which, if satisfied, makes the total energy of the global minima, as well as the main properties of the model, equivalent to those obtained for the $G=SU(2)$ case. We study this condition and its consequences within the generalized rational map ansatz and show how it can be satisfied for $G=SU(p+q)$, where $p$ and $q$ are positive integers, with the Hermitian symmetric spaces being $SU(p+q)/SU(p) \otimes SU(q) \otimes U(1)$. In such a case, the model is completely equivalent to the $G=SU(2)$ False Vacuum Skyrme model, independent of the values of $p$ and $q$. We also provide a numerical study of the baryon density, RMS radius, and binding energy per nucleon, which deepens the previous analysis conducted for the $SU(2)$ False Vacuum Skyrme model. Additionaly, in the case of $G = SU(3)$, we have studied the application of our model to the description of the binding energies and masses of the $\Lambda$-hypernuclei.

We consider the scaling behaviour of the fluctuation susceptibility associated with the average activation in the Greenberg-Hastings neural network model and its relation to microscopic spontaneous activation. We found that, as the spontaneous activation probability tends to zero, a clear finite size scaling behaviour in the susceptibility emerges, characterized by critical exponents which follow already known scaling laws. This shows that the spontaneous activation probability plays the role of an external field conjugated to the order parameter of the dynamical activation transition. The roles of different kinds of activation mechanisms around the different dynamical phase transitions exhibited by the model are characterized numerically and using a mean field approximation.

The nonlinear and chaotic nature of turbulent flows poses a major challenge for designing effective control strategies to maintain or induce low-drag laminar states. Traditional linear methods often fail to capture the complex dynamics governing transitions between laminar and turbulent regimes. In this work, we introduce the concept of the minimal seed for relaminarization-the closest point to a reference state in the turbulent region of the state space that triggers a direct transition to laminar flow without a chaotic transient. We formulate the identification of this optimal perturbation as a fully nonlinear optimization problem and develop a numerical framework based on a multi-step penalty method to compute it. Applying this framework to a nine-mode model of a sinusoidal shear flow, we compute the minimal seeds for both transition to turbulence and relaminarization. While both of these minimal seeds lie infinitesimally close to the laminar-turbulent boundary-the edge of chaos-they are generally unrelated and lie in distant and qualitatively distinct regions of state space, thereby providing different insights into the flow's underlying structure. We find that the optimal perturbation for triggering transition is primarily in the direction of the mode representing streamwise vortices (rolls), whereas the optimal perturbation for relaminarization is distributed across multiple modes without strong contributions in the roll or streak directions. By analyzing trajectories originating from these minimal seeds, we find that both transition and laminarization behavior are controlled by the stable and unstable manifolds of a periodic orbit on the edge of chaos. The laminarizing trajectory obtained from the minimal seed for relaminarization provides an efficient pathway out of turbulence and can inform the design and evaluation of flow control strategies aimed at inducing laminarization.

The classical dynamical $r$-matrix structure for the periodic elliptic Ruijsenaars chain is described. The Poisson brackets for the monodromy matrix are calculated as well, thus providing Liouville integrability of the model. Next, we study its continuous non-relativistic limit and reproduce the Maillet type non-ultralocal $r$-matrix structure for the field analogue of the elliptic Calogero-Moser model.