Fluctuation-Response Theory for Nonequilibrium Langevin Dynamics
Hyun-Myung Chun, Euijoon Kwon, Hyunggyu Park, Jae Sung Lee · 2026
Read Fluctuation-Response Theory for Nonequilibrium Langevin Dynamics by Hyun-Myung Chun, Euijoon Kwon, Hyunggyu Park, Jae Sung Lee (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-16387.txt.
Introduction
Hyun-Myung Chun and three coauthors set one law for waver and reply in Langevin flows. Their 2026 preprint deals with flows bent by chance heat noise. Past laws link waver size to reply size at rest. Far from rest, that old link no longer holds in plain form. The team seeks one law that holds in both zones.
They use shared maths in the waver of mass and flow counts. From it they draw one tie that spans rest and non rest. A wide law lets teams read drive from waver in real flows. It also tests if a flow is at rest from data. The work firms up non rest flow maths.
Catalog record
| CREATOR | Hyun-Myung Chun, Euijoon Kwon, Hyunggyu Park, Jae Sung Lee |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 38 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2601-16387
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Fluctuation-Response Theory for Nonequilibrium Langevin Dynamics Hyun-Myung Chun, Euijoon Kwon, Hyunggyu Park, Jae Sung Lee Abstract We establish a unified fluctuation–response relation for Langevin dynamics. By exploiting the common mathematical structures underlying fluctuations and responses of empirical density and current, we derive a unified identity that generalizes the fluctuation–dissipation theorem from equilibrium to nonequilibrium settings. This relation connects global fluctuations of observables with their local responses to perturbations in force, mobility, and temperature. We further derive finite-time fluctuation–response inequalities, leading to response uncertainty relations that complement the identity by providing more practical bounds. These derivations establish a unified theoretical framework linking the fluctuation–dissipation theorem and thermodynamic uncertainty relations. Using the F 1 F_{1} -ATPase molecular motor model, we illustrate how these response-based bounds constrain the long-time diffusion coefficient. Introduction – Understanding the relationship between fluctuations and response is a central theme in statistical physics. Near equilibrium, this relationship is firmly established by the seminal fluctuation-dissipation theorem (FDT), which expresses the linear response of a system in terms of its equilibrium fluctuations Kubo 1966 ; Kubo et al. 2012 ; Marconi et al. 2008 . However, far from equilibrium, this relation no longer holds in its original form, motivating efforts to uncover generalized principles. One important direction has been the extension of the FDT to nonequilibrium systems Agarwal 1972 ; Speck and Seifert 2006 ; Baiesi et al. 2009 ; Prost et al. 2009 ; Seifert and Speck 2010 ; Altaner et al. 2016 , where dissipation and dynamical activity play central roles in characterizing the response Baiesi et al. 2009 ; Seifert and Speck 2010 . In parallel, thermodynamic and kinetic uncertainty relations (TUR and KUR) have revealed fundamental bounds on current fluctuations, highlighting intrinsic trade-offs between fluctuation and dissipation and between fluctuation and dynamical activity Barato and Seifert 2015 ; Gingrich et al. 2016 ; Garrahan 2017 ; Horowitz and Gingrich 2020 ; Van Vu et al. 2022 ; Kwon et al. 2024 ; Lee et al. 2021 . These two lines of research on nonequilibrium fluctuations have advanced independently, without a clear link between them. More recently, the response-TUR (R-TUR) was proposed as a conjecture for Markov jump processes Ptaszyński et al. 2024 , constraining the response of currents in terms of fluctuations and dissipation. This conjecture was subsequently proven in the long-time limit through the discovery of fluctuation-response relations (FRRs) Aslyamov et al. 2025a , identities that connect fluctuations with nonequilibrium responses to local perturbations of system parameters. The R-TUR then follows as a corollary of these identities. Later, fluctuation-response inequalities (FRIs) were formulated, sharing the same structural form as the FRRs but appearing as finite-time inequalities that converge asymptotically to the FRRs Kwon et al. 2025 . Complementary studies, including the response kinetic uncertainty relation (R-KUR) Liu and Gu 2025 and quantum generalizations Vu 2025 ; Kwon et al. 2025 , have further broadened this framework. Yet, existing formulations of the FRRs and FRIs have been largely confined to Markov jump processes. Thus, although extending these relations to Langevin dynamics, the central paradigm for continuous-state nonequilibrium systems, has been repeatedly highlighted as an important direction Aslyamov et al. 2025a ; Ptaszynski et al. 2024 ; Ptaszynski et al. 2025 ; Liu and Gu 2025 , it has remained largely unexplored. In this Letter, we establish a unified FRR for nonequilibrium Langevin dynamics in the long-time limit. This relation arises from a structural correspondence between fluctuations and responses to local perturbations of empirical density and current. By incorporating the local Onsager reciprocal relation that we derive in equilibrium, we demonstrate that the FRR reduces to the FDT in the equilibrium limit. Moreover, using the Cramér-Rao bound, we derive finite-time FRI that extends the connection between fluctuations and responses beyond the long-time limit and to arbitrary initial conditions. From the FRI, we further obtain response uncertainty relations, which provide experimentally accessible formulations based on global perturbations, as we shall demonstrate with the F_{1} -ATPase molecular motor model. These derivations complete the overall hierarchical relationship among the FRI, FRR, FDT, and TUR. Setup – We formulate the main results in one-dimensional overdamped Langevin systems for clarity of presentation. This simple case already captures the essence of the FRR and FRI while avoiding technical complications. The extension to multidimensional settings is presented in Ref. SM . Thus, we consider a one-dimensional overdamped Langevin system described by \dot{x}_{t}=\mu(x_{t})F(x_{t})+\sqrt{2\mu(x_{t})T(x_{t})}\circledast\xi_{t}\ , (1) where \circledast denotes the anti-Itô product Lau and Lubensky 2007 ; ant , F(x) is an external force, and \xi_{t} is a zero-mean Gaussian white noise with \langle\xi_{t}\xi_{s}\rangle=\delta(t-s) . \mu(x) and T(x) represent the position-dependent mobility and bath temperature, respectively. We set the Boltzmann constant to unity throughout. The corresponding Fokker-Planck equation for the propagator P(x,t|y,s) is given by \partial_{t}P(x,t|y,s)=\hat{\mathcal{L}}_{x}P(x,t|y,s), (2) where \hat{\mathcal{L}}_{x}\equiv-\partial_{x}\mu(x)[F(x)-T(x)\partial_{x}] . Hereafter, the differential operators, such as \hat{\mathcal{L}}_{x} and \partial_{x} , apply to all functions to their right, unless parentheses are used to restrict their range. For instance, \partial_{x}f(x)g(x)=[\partial_{x}f(x)]g(x)+[\partial_{x}g(x)]f(x) . For notational convenience, we introduce the differential operator \hat{\mathcal{J}}_{x}\equiv\mu(x)[F(x)-T(x)\partial_{x}] , so that \hat{\mathcal{L}}_{x}=-\partial_{x}\hat{\mathcal{J}}_{x} . We consider the following time-averaged observable: \Theta(\tau)=\frac{1}{\tau}\int_{0}^{\tau}[f(x_{t})+\dot{x}_{t}\circ g(x_{t})]dt (3) with arbitrary functions f(x) and g(x) , where \circ denotes the Stratonovich product Gardiner 1985 . Its steady-state average is given by \langle\Theta\rangle_{\rm ss}=\int[f(x)\pi(x)+g(x)j_{\rm ss}(x)]dx , where \pi(x) represents the steady-state probability distribution satisfying \hat{\mathcal{L}}_{x}\pi(x)=0 , and j_{\rm ss}(x)\equiv\hat{\mathcal{J}}_{x}\pi(x) is the steady-state probability current 1 1 1 Although the steady-state current j_{\rm ss}(x) is spatially constant in the one-dimensional case, we keep the argument explicit to facilitate a clear connection to more general cases. . Fluctuation-response relations – We consider first the response of \langle\Theta\rangle_{\rm ss} to a local perturbation in a function \phi(x) at x=z , expressed as the functional derivative \delta\langle\Theta\rangle_{\rm ss}/\delta\phi(z) . We restrict \phi to the set \{\mu,F,T\} , as these cases share a common structural form for the response function. A local perturbation in \phi at position z amounts to a shift of the differential operator \hat{\mathcal{J}}_{x} into the form of \hat{\mathcal{J}}_{x}+\varepsilon\delta(x-z)\hat{\mathcal{K}}_{x}^{\phi} , where \varepsilon is an infinitesimal factor and \hat{\mathcal{K}}_{x}^{\phi} is an operator determined by the choice of \phi (see Table 1 ). For two observables \Theta_{1} and \Theta_{2} , our first main result–the FRR–relates the scaled covariance C_{\Theta_{1},\Theta_{2}}\equiv\lim_{\tau\to\infty}\tau{\rm Cov}\{\Theta_{1}(\tau),\Theta_{2}(\tau)\} to the responses associated with two functions \phi_{1} and \phi_{2} through the identity C_{\Theta_{1},\Theta_{2}}=\int\frac{2\pi(z)D(z)}{N_{\phi_{1}}(z)N_{\phi_{2}}(z)}\frac{\delta\langle\Theta_{1}\rangle_{\rm ss}}{\delta\phi_{1}(z)}\frac{\delta\langle\Theta_{2}\rangle_{\rm ss}}{\delta\phi_{2}(z)}dz\;, (4) where D(z)\equiv\mu(z)T(z) is the diffusion coefficient and N_{\phi}(x)\equiv\hat{\mathcal{K}}_{x}^{\phi}\pi(x) (see Table 1 ). This relation represents the continuum (Langevin) counterpart of the FRRs established for Markov jump processes Aslyamov et al. 2025a . This single identity holds for all cross-combinations of \phi_{1} and \phi_{2}\in\{\mu,F,T\} . It thus provides a generalization of the FDT, connecting global fluctuations of observables to their local responses in nonequilibrium Langevin systems. \phi \hat{\mathcal{K}}_{x}^{\phi} N_{\phi}(x)=\hat{\mathcal{K}}_{x}^{\phi}\pi(x) \mu F(x)-T(x)\,\partial_{x} j_{\rm ss}(x)/\mu(x) F \mu(x) \mu(x)\,\pi(x) T -\mu(x)\,\partial_{x} -\mu(x)\,\partial_{x}\pi(x) Table 1: Perturbation operators \hat{\mathcal{K}}_{x}^{\phi} and prefactors \mathcal{N}_{\phi}(x) for different choices of the perturbed function \phi\in\{\mu,F,T\} . A brief outline of the derivation of Eq. ( 4 ) is given below. Linear response analysis yields the following expressions (see Appendix A): \displaystyle\frac{\delta\pi(x)}{\delta\phi(z)}=N_{\phi}(z)\partial_{z}H(x|z),\quad\frac{\delta j_{\rm ss}(x)}{\delta\phi(z)}=N_{\phi}(z)R(x,z) (5) with H(x|z)\equiv\int_{0}^{\infty}[P(x,t|z,0)-\pi(x)]dt and R(x,z)\equiv\delta(x-z)+\hat{\mathcal{J}}_{x}\partial_{z}H(x|z) . Since the factor N_{\phi}(z) is shared by both the response of \pi(x) and that of j_{\rm ss}(x) , the ratio of the local response