The entropy production of stationary diffusions
Open accessThe DOI matches its Crossref record; 60 of the 63 references it cites were found in a registry or an index.
Abstract
The entropy production rate is a central quantity in non-equilibrium statistical physics, scoring how far a stochastic process is from being time-reversible. In this paper, we compute the entropy production of diffusion processes at non-equilibrium steady-state under the condition that the time-reversal of the diffusion remains a diffusion. We start by characterising the entropy production of both discrete and continuous-time Markov processes. We investigate the time-reversal of time-homogeneous stationary diffusions and recall the most general conditions for the reversibility of the diffusion property, which includes hypoelliptic and degenerate diffusions, and locally Lipschitz vector fields. We decompose the drift into its time-reversible and irreversible parts, or equivalently, the generator into symmetric and antisymmetric operators. We show the equivalence with a decomposition of the backward Kolmogorov equation considered in hypocoercivity theory, and a decomposition of the Fokker-Planck equation in GENERIC form. The main result shows that when the time-irreversible part of the drift is in the range of the volatility matrix (almost everywhere) the forward and time-reversed path space measures of the process are mutually equivalent, and evaluates the entropy production. When this does not hold, the measures are mutually singular and the entropy production is infinite. We verify these results using exact numerical simulations of linear diffusions. We illustrate the discrepancy between the entropy production of non-linear diffusions and their numerical simulations in several examples and illustrate how the entropy production can be used for accurate numerical simulation. Finally, we discuss the relationship between time-irreversibility and sampling efficiency, and how we can modify the definition of entropy production to score how far a process is from being generalised reversible.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“This means precise particles exhibit solenoidal behaviour such as oscillatory and (quasi) periodic orbits—and an accompanying loss of detailed balance, i.e., turbulent and time-irreversible dynamics [106, 107, 30].”
From The free energy principle made simpler but not too simple (2023) · cites this paper as [30]
Citations
Cites
On the Theory of the Brownian Motion
Uhlenbeck, G. E., & Ornstein, L. S. (1930) · Physical Review · doi:10.1103/physrev.36.823 · Cited by 4,482
Stochastic thermodynamics, fluctuation theorems and molecular machines
Seifert, U. (2012) · Reports on Progress in Physics · doi:10.1088/0034-4885/75/12/126001 · Cited by 3,300 · cited 1×
“important tool in biophysics, as a measure of the metabolic cost of molecular processes, such as molecular motors [2], and it was shown empirically that brain states associated with effortful activity correlated with a higher entropy production from neural activity [6].”
MethodDynamical Ensembles in Nonequilibrium Statistical Mechanics
Gallavotti, G., & Cohen, E. G. D. (1995) · Physical Review Letters · doi:10.1103/physrevlett.74.2694 · Cited by 1,703
A Gallavotti–Cohen-Type Symmetry in the Large Deviation Functional for Stochastic Dynamics
Lebowitz, J. L., & Spohn, H. (1999) · Journal of Statistical Physics · doi:10.1023/a:1004589714161 · Cited by 1,417
Stochastic boundary conditions for molecular dynamics simulations of ST2 water
Brünger, A., Brooks, C. L., & Karplus, M. (1984) · Chemical Physics Letters · doi:10.1016/0009-2614(84)80098-6 · Cited by 643 · cited 13×
“2 BBK discretisation Contrariwise to Euler, the BBK integrator [26, 27] is a splitting scheme that when applied to underdamped Langevin yields absolutely continuous transition kernels.”
BackgroundMethodInfluentialErgodicity for SDEs and approximations: locally Lipschitz vector fields and degenerate noise
Mattingly, J. C., Stuart, A. M., & Higham, D. J. (2002) · Stochastic Processes and their Applications · doi:10.1016/s0304-4149(02)00150-3 · Cited by 550
Potential in stochastic differential equations: novel construction
Ao, P. (2004) · Journal of Physics A: Mathematical and General · doi:10.1088/0305-4470/37/3/l01 · Cited by 261
Is there an analog of Nesterov acceleration for gradient-based MCMC?
Ma, Y.-A., Chatterji, N. S., Cheng, X., Flammarion, N., Bartlett, P. L., & Jordan, M. I. (2021) · Bernoulli · doi:10.3150/20-bej1297 · Cited by 215 · cited 2×
“Underdamped Langevin dynamics is an important model in statistical physics and sampling [10,27,45].”
BackgroundMethodChaudhari, P., & Soatto, S. (2018) · 2018 Information Theory and Applications Workshop (ITA) · doi:10.1109/ita.2018.8503224 · Cited by 213
Haussmann, U. G., & Pardoux, E. (1986) · The Annals of Probability · doi:10.1214/aop/1176992362 · Cited by 210 · cited 1×
“In general, the timereversal of a diffusion need not be a diffusion [17], but Haussman and Pardoux showed that the diffusion property is preserved under some mild regularity conditions on the diffusion process [33].”
BackgroundCovariant formulation of non-equilibrium statistical thermodynamics
Graham, R. (1977) · Zeitschrift für Physik B Condensed Matter and Quanta · doi:10.1007/bf01570750 · Cited by 203 · cited 2×
“"⇒" Recall that any smooth divergence-free vector field is the divergence of a smooth antisymmetric matrix field A = −A> [7, 8, 71] birrρ = ∇ ·A.”
BackgroundConvergence of Numerical Time-Averaging and Stationary Measures via Poisson Equations
Mattingly, J. C., Stuart, A., & Tretyakov, M. (2009) · SIAM Journal on Numerical Analysis · doi:10.1137/090770527 · Cited by 191
Multidimensional thermodynamic uncertainty relations
Dechant, A. (2018) · Journal of Physics A Mathematical and Theoretical · doi:10.1088/1751-8121/aaf3ff · Cited by 185 · cited 1×
“More recently, entropy production is at the heart of the so-called thermodynamic uncertainty relations, which provide estimates of ep from observations of the system [5].”
BackgroundA Complete Recipe for Stochastic Gradient MCMC
Ma, Y.-A., Chen, T., & Fox, E. B. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1506.04696 · Cited by 175
Broken detailed balance and entropy production in the human brain
Lynn, C. W., Cornblath, E. J., Papadopoulos, L., Bertolero, M. A., & Bassett, D. S. (2021) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2109889118 · Cited by 163 · cited 1×
“important tool in biophysics, as a measure of the metabolic cost of molecular processes, such as molecular motors [2], and it was shown empirically that brain states associated with effortful activity correlated with a higher entropy production from neural activity [6].”
MethodVariance Reduction Using Nonreversible Langevin Samplers
Duncan, A., Lelièvre, T., & Pavliotis, G. A. (2016) · Journal of Statistical Physics · doi:10.1007/s10955-016-1491-2 · Cited by 155 · cited 3×
“1 ep and sampling efficiency A well-known criterion for efficient sampling is time-irreversibility [13, 20, 21, 64].”
BackgroundEntropy Production in Nonlinear, Thermally Driven Hamiltonian Systems
Eckmann, J., Pillet, C., & Rey-Bellet, L. (1999) · Journal of Statistical Physics · doi:10.1023/a:1004537730090 · Cited by 130
Chafaï, D. (2004) · Kyoto journal of mathematics · doi:10.1215/kjm/1250283556 · Cited by 129
Irreversible Langevin samplers and variance reduction: a large deviations approach
Rey-Bellet, L., & Spiliopoulos, K. (2015) · Nonlinearity · doi:10.1088/0951-7715/28/7/2081 · Cited by 126 · cited 3×
“1 ep and sampling efficiency A well-known criterion for efficient sampling is time-irreversibility [13, 20, 21, 64].”
BackgroundHydrodynamics and fluctuations outside of local equilibrium: Driven diffusive systems
Eyink, G. L., Lebowitz, J. L., & Spohn, H. E. (1996) · Journal of Statistical Physics · doi:10.1007/bf02183738 · Cited by 126 · cited 1×
“"⇒" Recall that any smooth divergence-free vector field is the divergence of a smooth antisymmetric matrix field A = −A> [7, 8, 71] birrρ = ∇ ·A.”
BackgroundOptimal non-reversible linear drift for the convergence to equilibrium of a diffusion
Lelièvre, T., Nier, F., & Pavliotis, G. A. (2013) · Journal of Statistical Physics · doi:10.1007/s10955-013-0769-x · Cited by 123 · cited 1×
“In elliptic linear diffusions, one can construct the optimal time-irreversible drift to optimise the spectral gap [50, 66] or the asymptotic variance [64].”
BackgroundSelf-citationFöllmer, H. (1986) · Lecture notes in mathematics · doi:10.1007/bfb0080212 · Cited by 121 · cited 1×
“Similar time-reversal theorems exist in various settings: for more singular coefficients on the torus [36], under (entropic) regularity conditions on the forward path space measure [37,38], for infinite-dimensional diffusions [39–41], for diffusions on open time-intervals [42], or with boundary conditions [43].”
BackgroundNonequilibrium Thermodynamics of Stochastic Systems with Odd and Even Variables
Spinney, R. E., & Ford, I. J. (2012) · Physical Review Letters · doi:10.1103/physrevlett.108.170603 · Cited by 120
Equivalence of Stochastic Equations and Martingale Problems
Kurtz, T. G. (2010) · Springer eBooks · doi:10.1007/978-3-642-15358-7_6 · Cited by 113
Hwang, C.-R., Hwang-Ma, S.-Y., & Sheu, S.-J. (2005) · The Annals of Applied Probability · doi:10.1214/105051605000000025 · Cited by 113 · cited 2×
“1 ep and sampling efficiency A well-known criterion for efficient sampling is time-irreversibility [13, 20, 21, 64].”
BackgroundOn Malliavinʼs proof of Hörmanderʼs theorem
Hairer, M. (2011) · Bulletin des Sciences Mathématiques · doi:10.1016/j.bulsci.2011.07.007 · Cited by 99
Uncertainty relations for underdamped Langevin dynamics
Vu, T. V., & Hasegawa, Y. (2019) · Physical review. E · doi:10.1103/physreve.100.032130 · Cited by 96
Large deviations and applications
Varadhan, S. R. S. (1988) · Lecture notes in mathematics · doi:10.1007/bfb0086178 · Cited by 90
Integration by Parts and Time Reversal for Diffusion Processes
Millet, A., Nualart, D., & Sanz, M. (1989) · The Annals of Probability · doi:10.1214/aop/1176991505 · Cited by 81 · cited 5×
“In our treatment, we only assume that the time-reversal of a diffusion is a diffusion (the necessary and sufficient conditions for which were first established by Millet, Nualart and Sanz [17]) and sufficient regularity to apply Girsanov’s theorem or the Stroock-Varadhan support theorem.”
BackgroundMethodResultInfluentialCharacterising the nonequilibrium stationary states of Ornstein–Uhlenbeck processes
Godrèche, C., & Luck, J. M. (2018) · Journal of Physics A Mathematical and Theoretical · doi:10.1088/1751-8121/aaf190 · Cited by 64 · cited 1×
“, the one dimensional ferromagnetic Gaussian spin model [55]); or when one linearises the equations of generic diffusion processes near the stationary density.”
BackgroundBayesian mechanics for stationary processes
Da Costa, L., Friston, K., Heins, C., & Pavliotis, G. A. (2021) · Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences · doi:10.1098/rspa.2021.0518 · Cited by 63
Self-citationUsing Perturbed Underdamped Langevin Dynamics to Efficiently Sample from Probability Distributions
Duncan, A., Nüsken, N., & Pavliotis, G. A. (2017) · Journal of Statistical Physics · doi:10.1007/s10955-017-1906-8 · Cited by 58
Stochastic Chaos and Markov Blankets
Friston, K., Heins, C., Ueltzhöffer, K., Da Costa, L., & Parr, T. (2021) · Entropy · doi:10.3390/e23091220 · Cited by 56
Rey-Bellet, L. (2006) · Lecture notes in mathematics · doi:10.1007/3-540-33966-3_2 · Cited by 54 · cited 1×
“This process arises in statistical physics, as a model of a particle coupled to a heat bath [59], [10, Chapter 8]; in Markov chain Monte-Carlo as an accelerated sampling scheme [13,45]; and also as a model of interacting kinetic particles.”
MethodA decomposition of irreversible diffusion processes without detailed balance
Qian, H. (2013) · Journal of Mathematical Physics · doi:10.1063/1.4803847 · Cited by 51
Improving Asymptotic Variance of MCMC Estimators: Non-reversible Chains are BetterPreprint
Neal, R. M. (2004) · arXiv (Cornell University) · doi:10.48550/arxiv.math/0407281 · Cited by 48 · cited 1×
“Intuitively, non-reversible processes backtrack less often and thus furnish more diverse samples [65].”
BackgroundTime reversal of infinite-dimensional diffusions
Föllmer, H., & Wakolbinger, A. (1986) · Stochastic Processes and their Applications · doi:10.1016/0304-4149(86)90114-6 · Cited by 48
Markov Chain Monte Carlo and Irreversibility
Ottobre, M. (2016) · Reports on Mathematical Physics · doi:10.1016/s0034-4877(16)30031-3 · Cited by 43
Dynamical large deviations of linear diffusions
du Buisson, J., & Touchette, H. (2023) · Physical review. E · doi:10.1103/physreve.107.054111 · Cited by 40
Optimal linear drift for the speed of convergence of an hypoelliptic diffusion
Guillin, A., & Monmarché, P. (2016) · Electronic Communications in Probability · doi:10.1214/16-ecp25 · Cited by 37
A new proof for the conditions of Novikov and Kazamaki
Ruf, J. (2012) · Stochastic Processes and their Applications · doi:10.1016/j.spa.2012.09.011 · Cited by 32
Geometric methods for sampling, optimization, inference, and adaptive agents
Barp, A., Costa, L. D., França, G., Friston, K. J., Girolami, M., Jordan, M. I., & Pavliotis, G. A. (2022) · Handbook of statistics · doi:10.1016/bs.host.2022.03.005 · Cited by 32
Attaining the Optimal Gaussian Diffusion Acceleration
Wu, S., Hwang, C.-R., & Chu, M. T. (2014) · Journal of Statistical Physics · doi:10.1007/s10955-014-0963-5 · Cited by 27 · cited 1×
“In elliptic linear diffusions, one can construct the optimal time-irreversible drift to optimise the spectral gap [50, 66] or the asymptotic variance [64].”
BackgroundMazzolo, A., & Monthus, C. (2023) · Physical review. E · doi:10.1103/physreve.107.014101 · Cited by 25
A Unifying and Canonical Description of Measure-Preserving DiffusionsPreprint
Barp, A., Takao, S., Betancourt, M., Arnaudon, A., & Girolami, M. (2021) · Cambridge University Engineering Department Publications Database · doi:10.48550/arxiv.2105.02845 · Cited by 23 · cited 2×
“This decomposition is called the Helmholtz decomposition because it can be obtained geometrically by decomposing the drift vector field b into horizontal birr (time-irreversible, conservative) and vertical brev (time-reversible, non-conservative) components with respect to the stationary density [11].”
MethodTime reversal for infinite-dimensional diffusions
Millet, A., Nualart, D., & Sanz, M. (1989) · Probability Theory and Related Fields · doi:10.1007/bf00339991 · Cited by 22
Time reversal of diffusion processes under a finite entropy condition
Cattiaux, P., Conforti, G., Ivan Gentil, C., & Léonard, C. (2023) · Annales de l Institut Henri Poincaré Probabilités et Statistiques · doi:10.1214/22-aihp1320 · Cited by 15 · cited 1×
“Similar time-reversal theorems exist in various settings: for more singular coefficients on the torus [36], under (entropic) regularity conditions on the forward path space measure [37,38], for infinite-dimensional diffusions [39–41], for diffusions on open time-intervals [42], or with boundary conditions [43].”
BackgroundEntropy Production in Continuous Phase Space Systems
Luposchainsky, D., & Haye Hinrichsen, B. (2013) · Journal of Statistical Physics · doi:10.1007/s10955-013-0863-0 · Cited by 14
Yang, Y.-J., & Qian, H. (2021) · Journal of Statistical Physics · doi:10.1007/s10955-021-02723-3 · Cited by 13 · cited 1×
“"⇒" Recall that any smooth divergence-free vector field is the divergence of a smooth antisymmetric matrix field A = −A⊤ [7, 8, 71,72] birrρ = ∇ ·A.”
BackgroundPotentials of continuous Markov processes and random perturbations
Yang, Y.-J., & Cheng, Y.-C. (2021) · Journal of Physics A Mathematical and Theoretical · doi:10.1088/1751-8121/abef80 · Cited by 9
Time-reversal of diffusion processes and non-linear smoothing
Pardoux, É. (2006) · Springer eBooks · doi:10.1007/bfb0043398 · Cited by 7
Diffusion processes on an open time interval and their time reversal
Nagasawa, M., & Domenig, T. (1996) · Springer eBooks · doi:10.1007/978-4-431-68532-6_17 · Cited by 7 · cited 1×
“Similar time-reversal theorems exist in various settings: for more singular coefficients on the torus [36], under (entropic) regularity conditions on the forward path space measure [37,38], for infinite-dimensional diffusions [39–41], for diffusions on open time-intervals [42], or with boundary conditions [43].”
BackgroundTime reversal of diffusion processes with a boundary condition
Cattiaux, P. (1988) · Stochastic Processes and their Applications · doi:10.1016/0304-4149(88)90101-9 · Cited by 6 · cited 1×
“Similar time-reversal theorems exist in various settings: for more singular coefficients on the torus [36], under (entropic) regularity conditions on the forward path space measure [37,38], for infinite-dimensional diffusions [39–41], for diffusions on open time-intervals [42], or with boundary conditions [43].”
BackgroundMeasuring the Irreversibility of Numerical Schemes for Reversible Stochastic Differential Equations
Katsoulakis, M., Pantazis, Y., & Rey-Bellet, L. (2013) · ESAIM Mathematical Modelling and Numerical Analysis · doi:10.1051/m2an/2013142 · Cited by 5 · cited 3×
“This definition is useful, for example, to quantify the entropy production of numerical simulations of a stochastic process [22].”
BackgroundInfluentialMeasuring the Irreversibility of Numerical Schemes for Reversible Stochastic Differential EquationsPreprint
Katsoulakis, M. A., Pantazis, Y., & Rey-Bellet, L. (2012) · ACMAC's PrePrint Repository (Archimedes Center for Modeling, Analysis and Computation) · doi:10.1051/m2an/2013142/pdf · Cited by 5
Non-reversible processes: GENERIC, hypocoercivity and fluctuations
Duong, M. H., & Ottobre, M. (2023) · Nonlinearity · doi:10.1088/1361-6544/acb47b · Cited by 4
Time Reversal of Diffusion Processes in Hilbert Spaces and Manifolds
Belopolskaya, Y. I. (2001) · Birkhäuser Boston eBooks · doi:10.1007/978-1-4612-0209-7_6 · Cited by 2
Time Reversal of Degenerate Diffusions
Quastel, J. (2002) · Birkhäuser Boston eBooks · doi:10.1007/978-1-4612-0063-5_10 · Cited by 2 · cited 1×
“Similar time-reversal theorems exist in various settings: for more singular coefficients on the torus [36], under (entropic) regularity conditions on the forward path space measure [37,38], for infinite-dimensional diffusions [39–41], for diffusions on open time-intervals [42], or with boundary conditions [43].”
BackgroundTopological support of solutions to stochastic differential equations
Šimon, P. (2016) · Digital Repository (National Repository of Grey Literature) · Cited by 1
Time reversal of diffusion processes under a finite entropy conditionPreprint
Cattiaux, P., Conforti, G., Gentil, I., et al. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2104.07708 · Cited by 1 · cited 1×
“Similar time-reversal theorems exist in various settings: for more singular coefficients on the torus [36], under (entropic) regularity conditions on the forward path space measure [37,38], for infinite-dimensional diffusions [39–41], for diffusions on open time-intervals [42], or with boundary conditions [43].”
Background
Research works printed but not linked: 1
- Talay, 2002, Stochastic hamiltonian systems: exponential convergence to the invariant measure and discretization by the implicit euler scheme, Markov Process. Relat. Fields, 8, 163
Cited by
The free energy principle made simpler but not too simple
Friston, K., Da Costa, L., Sajid, N., Heins, C., Ueltzhöffer, K., Pavliotis, G. A., & Parr, T. (2023) · Physics Reports · doi:10.1016/j.physrep.2023.07.001 · Cited by 163
“The final term, Λ , can be regarded as a correction term, which is neither curl-free nor divergence-free, and which ensures that the probability density remains constant over time [30].”
Nonequilibrium physics of brain dynamics
Nartallo-Kaluarachchi, R., Kringelbach, M. L., Deco, G., Lambiotte, R., & Goriely, A. (2025) · Physics Reports · doi:10.1016/j.physrep.2025.10.003 · Cited by 35
“In addition, 𝑭 irr is odd whilst 𝑭 rev is even under time-reversal [122].”
Momani, S., & Ibrahim, R. W. (2025) · Frontiers in Applied Mathematics and Statistics · doi:10.3389/fams.2025.1643121 · Cited by 13
Form and function in biological filaments: a physicist’s review
Cammann, J., Laeverenz-Schlogelhofer, H., Wan, K. Y., & Mazza, M. G. (2025) · Philosophical Transactions of the Royal Society A Mathematical Physical and Engineering Sciences · doi:10.1098/rsta.2024.0253 · Cited by 19
Entropy production rate and time-reversibility for general jump diffusions on RRn
Zhang, Q., & Lu, Y. (2025) · Chaos An Interdisciplinary Journal of Nonlinear Science · doi:10.1063/5.0293862 · Cited by 4
“Lebowitz and Spohn [20] established a path-space framework connecting entropy production to large deviation theory and fluctuation theorems.”
Metadata
- Authors
- Da Costa, L., & Pavliotis, G. A.
- Affiliations
- Imperial College London (GB)
- Venue
- Journal of Physics A Mathematical and Theoretical
- Publisher
- IOP Publishing
- Published
- 2023-08-22
- Volume
- 56
- Issue
- 36
- Pages
- 365001
- ISSN
- 1751-8113 (ISSN-L)
- 1751-8121
- OpenAlex
- W4381162055
- Cited by
- 9
- References
- 63
- Citations by year
- 2026: 4 · 2025: 4 · 2023: 1
- FWCI
- 0.89
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- creativecommons.org/licenses/by/4.0 (vor)
- iopscience.iop.org/info/page/text-and-data-mining (tdm)
- Locations
- Journal of Physics A Mathematical and Theoretical (publishedVersion, cc-by)
- arXiv (Cornell University) (submittedVersion)
- Spiral (Imperial College London) (submittedVersion, cc-by)
- arXiv (Cornell University)
- Funders
- JPMorgan Chase & Co
- Engineering and Physical Sciences Research Council (EP/P031587/1)
- Fonds National de la Recherche Luxembourg (13568875)
- Topics
- Physical Sciences › Physics and Astronomy › Statistical and Nonlinear Physics › Advanced Thermodynamics and Statistical Mechanics
- Physical Sciences › Mathematics › Mathematical Physics › Stochastic processes and statistical mechanics
- Physical Sciences › Mathematics › Statistics and Probability › Markov Chains and Monte Carlo Methods
- Keywords
- entropy production, stationary diffusion, time reversal, nonequilibrium steady states, hypoelliptic diffusion, degenerate diffusion, backward Kolmogorov equation, hypocoercivity, Fokker-Planck equation, GENERIC formalism, measurement invariance, Markov processes, sampling efficiency, linear diffusion
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.