REVIEW 3 major objections 5 minor 2 cited by
Sharp Page transitions in generic Hamiltonian dynamics
T0 review · 3 major / 5 minor · reviewed 2026-08-09 · deepseek-v4-flash
Pith's one-line read Under generic chaotic local Hamiltonian dynamics with energy as the only conserved quantity, the min-entropy peak becomes a true cusp at a well-defined Page time, set by energy diffusion across the subsystem.
desk verdict Convincing Page-curve numerics in a non-integrable chain, but the sharp-cusp claim is extrapolated from Lindblad data that also fits a power law; worth review, not a slam dunk. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The central objects are (i) the entanglement Hamiltonian $H_A = -\log \rho_A$, whose low-energy spectrum determines the large eigenvalues of the reduced density matrix; (ii) the hydrodynamic ansatz $\rho_A(t) \propto \exp\big(-\int \beta(x,t)\, h(x)\, dx\big)$, with bond- and site-resolved inverse temperatures $\beta_i$ and a nonequilibrium current term; and (iii) the minimum gap $\min\{\ln \lambda_0 - \ln \lambda_1\}$ at a level crossing, which distinguishes a sharp transition (gap closes exponentially in system size) from a crossover (finite gap). The machinery works by tying each avoided crossing in the entanglement spectrum to a local sign change of $\beta(x,t)$: early crossings involve single local terms and remain open, while the Page-time crossing involves the whole subsystem flipping and closes exponentially. The Landau-Zener-like eigenvector swaps are quantified with the Bhattacharyya distance, and the Lindblad master equation provides a boundary-driven model in which larger systems ($M$ up to 12) can be simulated to test the scaling.
What would settle it
Compute the minimum gap $\min\{\ln \lambda_0 - \ln \lambda_1\}$ at the Page-time crossing in the Lindblad model for $M = 12$ and $M = 14$ with sufficiently fine time sweeps; if the gap saturates at a nonvanishing value rather than continuing to shrink with system size, the claimed exponential closure in the thermodynamic limit is falsified. A complementary check is to measure the $S_\infty$ peak in a cold-atom or superconducting implementation at $M \gtrsim 16$: a true cusp should keep sharpening as $M$ grows, whereas a crossover leaves a rounded maximum.
Extended reading notes
Core claim
The paper claims that in a subsystem $A$ of size $M$ initialized in the ceiling state (negative temperature) and coupled to a much larger cold bath, the min-entropy $S_\infty = -\log \lambda_{\max}$ evolves non-monotonically and its maximum becomes a genuine cusp as $M \to \infty$, at a time $t_{\mathrm{Page}}$. This sharpness comes from an isolated level crossing between the top two eigenvalues of $\rho_A$: before $t_{\mathrm{Page}}$ the dominant eigenvector is close to the ground state of $-H$, and after it, the ground state of $H$, which are macroscopically distinct. Numerical evidence from a Lindblad master-equation model with $M = 5,\dots,10$ shows the minimum gap at the Page time decreasing roughly exponentially with $M$ (while the earliest avoided crossing keeps a finite gap), and metastability of the ground state of $H$ as a sharp resonance before the Page time. The paper also constructs a hydrodynamic ansatz $\rho_A(t) \propto \exp\big(-\int \beta(x,t)\, h(x)\, dx\big)$ with a nonequilibrium current correction, verified against exact dynamics for $M=6$, $N=18$ and Lindblad simulations for $M=12$, which accounts for the level crossings and reproduces the Page time. This supports the conclusion that the Page transition is a first-order transition in the entanglement spectrum, and that the Page time is the time at which the effective inverse temperature of the bulk of $A$ changes sign.
Load-bearing premise
The quantitative claim of sharpness rests on extrapolating the minimum-gap scaling from Lindblad chains of 5 to 10 sites to the thermodynamic limit, where the data still allow a power-law decay and no size scaling exists for the full unitary dynamics.
Editorial extensions
If this is right
- The Page time is set by energy diffusion across region $A$, so in a diffusive system it should grow as $t_{\mathrm{Page}} \sim M^2$ with a system-dependent diffusion constant, rather than being set by scrambling.
- In generic one-dimensional systems without a finite-temperature phase transition, only the min-entropy ($\alpha = \infty$) sharpens; finite-$\alpha$ Rényi entropies should remain smooth unless the Hamiltonian has a finite-temperature transition.
- The sharp Page transition leaves local observables analytic: the cusp lives in the entanglement spectrum, so a phase transition appears in a quantity that local probes cannot see.
- The hydrodynamic ansatz gives a compact description of $\rho_A(t)$, and with randomized-measurement tools it can be used to reconstruct the entanglement Hamiltonian and locate the Page time in experiments with ancilla-based reset, as in superconducting qubit arrays.
- Boundary-driven Lindblad dynamics with only a few jump operators is enough to host a sharp Page transition, opening the study of entanglement transitions in dissipative spin chains.
Reading between the lines
- If the exponential gap scaling holds, the curvature of the cusp and the width of the metastable region should scale with the same length scale as the gap, giving a measurable exponent that distinguishes a first-order (exponential) from a continuous (power-law) entanglement transition.
- The hydrodynamic sign-change mechanism suggests a sharp experimental test: measure the local inverse-temperature profile $\beta(x,t)$ via local energy densities and check that the Page time coincides with the zero crossing at the site farthest from the bath; a mismatch would indicate the ansatz misses the mechanism.
- The gap-closure exponent may differ between unitary full-system dynamics and Lindblad dynamics because in the Lindblad case energy is lost at the boundary; checking whether $t_{\mathrm{Page}}$ still scales as $M^2$ in the unitary case with $N = 3M$ would clarify whether the Markovian bath modifies the transition.
- By analogy with the island formula, a sharp finite-$\alpha$ Page transition in a two-dimensional system with a finite-temperature phase transition would give a non-analytic von Neumann entropy at $\alpha = 1$, potentially realizing in a tabletop system the type of singularity the island formula predicts for black holes.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper studies the entanglement dynamics of a subsystem initialized in a high-energy (negative-temperature) state and coupled to a cold bath, in a nonintegrable mixed-field Ising chain with energy as the only conserved quantity. The authors observe a Page-curve-like non-monotonic evolution of Rényi entropies, identify a sharp peak in the min-entropy at a well-defined Page time, and propose that this peak becomes a cusp in the thermodynamic limit through an exponentially closing gap in the entanglement spectrum. They support this with exact Krylov evolution for small full-system sizes and with larger Lindblad simulations, and they introduce a hydrodynamic ansatz for the entanglement Hamiltonian with site- and bond-dependent inverse temperatures plus a current correction. The paper claims the Page transition is a first-order transition between macroscopically distinct states and discusses implications for quantum simulation and black-hole information ideas.
Significance. If the central claim holds, the paper extends Page-transition physics from integrable/charge-conserving models to generic nonintegrable local Hamiltonian dynamics, giving a concrete hydrodynamic mechanism for a sharp entanglement transition in a closed quantum system. The strength of the work is that the Page-curve behavior is observed in exact evolution of a small but genuinely nonintegrable system and in an independent Lindblad model, and the entanglement-Hamiltonian ansatz is tested against exact reduced density matrices. The paper also makes falsifiable predictions, for example that finite-index Rényi entropies should become sharp when the Hamiltonian supports a finite-temperature phase transition. However, the quantitative evidence for the thermodynamic-limit sharpness is limited to M=5–10 Lindblad data, and the paper itself concedes that a power-law decay is also consistent with those data; the full-system unitary dynamics, which the title and abstract emphasize, has no size scaling beyond a single M=6, N=18 case. The hydrodynamic explanation also relies on fitting the same parameters that are then used to explain the transition, so the explanatory claim is partly circular.
major comments (3)
- [§III B, Fig. 2] The central claim that the min-entropy peak sharpens to a cusp in the thermodynamic limit rests on the minimum gap min{ln λ0 − ln λ1} at the Page time decreasing with system size M in the Lindblad model. Fig. 2b shows data only for M=5,...,10 on a log scale, and the caption explicitly states that a power-law decay could also be consistent with the data due to the large error bars. Since the distinction between an exponential closing and a power-law closing is exactly what decides whether the transition is sharp (first-order) or merely a crossover, this extrapolation is load-bearing and currently underdetermined. The authors should either obtain data over a wider range of M, quantify the goodness of fit for exponential versus power-law forms, or provide a theoretical argument that fixes the scaling form.
- [§III (first paragraph) and §II A] The abstract and title claim that the cusp occurs in 'generic Hamiltonian dynamics,' but the quantitative finite-size scaling is performed only in the Lindblad model, while the full-system unitary dynamics is shown only for M=6, N=18. Section III states this explicitly: 'for full-system dynamics, we are unable to study a sufficiently wide range of system sizes to address quantitative size-dependence.' Because the Lindblad dissipator breaks energy conservation and imposes a zero-temperature boundary, an exponential gap in that model does not by itself establish the behavior in the energy-conserving unitary setting. The authors should present at least a preliminary size scaling for the full-system case (even a few additional M values, e.g., M=4,5,6,7 with fixed ratio N=3M), or clearly restrict the sharpness claim to the Lindblad model and provide a separate argument for why the unitary case should behave identically.
- [§IV, Eqs. (8)–(9)] The hydrodynamic explanation fits the free parameters β_i(t), β_{i,i+1}(t), and J^ZX_i(t) to the exact reduced density matrix, and then uses those same fitted profiles to account for the Page transition and to conjecture finite-α sharpness. As presented, this is a consistency check rather than a predictive derivation. To make the explanation load-bearing, the authors should provide an independent test: for example, obtain β(x,t) from an energy-diffusion equation (with the diffusion constant and boundary condition computed independently) and compare the predicted Page time and gap scaling with the exact results, or verify that the fitted profiles reproduce observables not used in the fit. Without such a test, the hydrodynamic mechanism remains a plausible interpretation rather than a demonstrated mechanism.
minor comments (5)
- [§II B] There is a typo near 'This approach allows us to simulation systems with M = 10, 12'; it should read 'to simulate systems.'
- [§III A, Fig. 1 caption] The figure caption says 'The vertical lines mark the time of the first crossover and the Page time' for panels (g) and (h), but the main text does not explicitly define how the first crossover time is located; a short definition would improve clarity.
- [§III B, Fig. 2] The error bars in Fig. 2 are stated to be estimated from the resolution of the sampling; the authors should also discuss possible systematic errors from the finite time-step of the Trotterized Lindblad evolution, especially for the smallest gaps.
- [§V] The last sentence of the abstract promises 'conditions under which the Page transition should remain sharp even for Rényi entropies of finite index α,' but the discussion in §V formulates this only as a future direction dependent on the existence of a finite-temperature phase transition. The abstract should be phrased as a conjecture or the corresponding condition should be stated more precisely.
- [Supplemental S5] There is a typo in the phrase 'our ansatz is indeeed a good approximation'; it should read 'indeed.'
Circularity Check
No circular derivation: the sharp Page-transition claim is anchored in direct, independently extracted entanglement-spectrum data, while the hydrodynamic ansatz is explicitly fitted and used descriptively; self-citations are background.
full rationale
The central claim that the min-entropy peak sharpens to a cusp at the Page time is established in Sec. III by direct numerical analysis of the exact dynamics: the largest eigenvalues of the reduced density matrix are tracked, the minimum gap at the avoided crossing is extracted as a function of M, and the eigenvector change is characterized via the Bhattacharyya distance. None of these observations requires the hydrodynamic ansatz, so the sharpness claim does not reduce to that ansatz. The hydrodynamic section (Sec. IV) is explicitly an interpretative layer: "In practice, since we are working with small systems, we determine the profile of β(x, t) by fitting the exact time-evolved state to the form of Eq. (8)." Because the β profiles are fitted in-sample, the statement that zeros of β coincide with the previously observed level crossings, and that the ansatz 'accounts for' the Page transition, is a descriptive mapping rather than an independent prediction of the transition; this is an interpretation and over-fitting caveat, not circularity in the sense of a prediction being equivalent to its input by construction. The paper itself flags the main evidence-quality limitations: Fig. 2's caption concedes "a power law decay could also be consistent with the data due to the large error bars", and Sec. III states that quantitative size-dependence is studied only in the Lindblad model, with only M=6, N=18 shown for full-system dynamics. These concessions weaken the thermodynamic-limit extrapolation for 'generic Hamiltonian dynamics', but they are robustness concerns, not circular steps. The self-citations ([16], [20], and the hydrodynamics reference [28] involving one author) are used as background for charge-conserved analogs or as a standard assumption; they are not load-bearing for the energy-only sharpness claim, and no external benchmark is contradicted. Therefore no circular step meets the quoted-reduction standard, and the appropriate score is 0.
Assumptions & free parameters
free parameters (4)
- Site inverse temperatures beta_i(t) =
time-dependent, fitted to exact rho_A(t)
- Bond inverse temperatures beta_{i,i+1}(t) =
time-dependent, fitted to exact rho_A(t)
- Current coefficients J^ZX_i(t) =
time-dependent, fitted to exact rho_A(t)
- Lindblad dissipation rate gamma =
unspecified
assumptions (6)
- domain assumption Mixed-field Ising chain at h = 0.809, g = 0.905, J = 1 is robustly nonintegrable.
- domain assumption rho_A(t) rapidly becomes locally thermal with a slowly varying inverse temperature beta(x,t).
- domain assumption Energy transport in the chain is diffusive and determines beta(x,t).
- domain assumption The Markovian Lindblad bath is a faithful stand-in for a large cold bath for the Page transition.
- domain assumption Ground states of H and -H are macroscopically distinct.
- domain assumption For highly excited eigenstates, rho_A is proportional to exp(-beta H) under ETH.
invented entities (1)
-
Nonequilibrium current operator J^ZX_i(t) sigma^z_i sigma^x_{i+1}
Cite this review
Pith. "Pith review of Sharp Page transitions in generic Hamiltonian dynamics." pith.science (2026). https://pith.science/paper/Z7JA65YF
@misc{pith2026250203524,
author = {Pith},
title = {Pith review of: Sharp Page transitions in generic Hamiltonian dynamics},
year = {2026},
howpublished = {\url{https://pith.science/paper/Z7JA65YF}},
note = {Machine review of arXiv:2502.03524}
}
abstract
We consider the entanglement dynamics of a subsystem initialized in a pure state at high energy density (corresponding to negative temperature) and coupled to a cold bath. The subsystem's R\'enyi entropies $S_\alpha$ first rise as the subsystem gets entangled with the bath and then fall as the subsystem cools. We find that the peak of the min-entropy, $\lim_{\alpha \to \infty} S_\alpha$, sharpens to a cusp in the thermodynamic limit at a well-defined time we call the Page time. We construct a hydrodynamic ansatz for the evolution of the entanglement Hamiltonian, which accounts for the sharp Page transition as well as the intricate dynamics of the entanglement spectrum before the Page time. Our results hold both when the bath has the same Hamiltonian as the system and when the bath is taken to be Markovian. Our ansatz suggests conditions under which the Page transition should remain sharp even for R\'enyi entropies of finite index $\alpha$.
Figures
Forward citations
Cited by 2 Pith papers
-
Page Curve and Entanglement Dynamics in an Interacting Fermionic Chain
In an interacting fermionic chain coupled to a reservoir, the entanglement entropy follows a Page curve and the min-entropy develops a non-analyticity whose thermodynamic-limit critical time vanishes as interactions grow.
-
Extreme dynamics and relaxation of quantum gases: A hydrodynamic approach
For a polytropic quantum gas, the paper claims a shock-front expansion with cloud size growing as a power of time, plus damped relaxation modes for a trapped gas; the vacuum shock result is physically problematic.
Reference graph
Works this paper leans on
-
[1]
J. Polchinski, in New Frontiers in Fields and Strings: TASI 2015 Proceedings of the 2015 Theoretical Advanced Study Institute in Elementary Particle Physics (World Scientific, 2017) pp. 353–397
work page 2015
-
[2]
A. Almheiri, D. Marolf, J. Polchinski, and J. Sully, Jour- nal of High Energy Physics 2013, 1 (2013)
work page 2013
-
[3]
Black Hole Complementarity and the Harlow-Hayden Conjecture
L. Susskind, arXiv preprint arXiv:1301.4505 (2013)
work page Pith review arXiv 2013
- [4]
- [5]
-
[6]
I. H. Kim and J. Preskill, Journal of High Energy Physics 2023, 1 (2023)
work page 2023
-
[7]
O. DeWolfe and K. Higginbotham, Journal of High En- ergy Physics 2023, 1 (2023)
work page 2023
-
[8]
R. Li, X. Wang, K. Zhang, and J. Wang, Phys. Rev. D 109, 044005 (2024)
work page 2024
Show all 36 references
-
[9]
Engelhardt, ˚A
N. Engelhardt, ˚A. Folkestad, A. Levine, E. Verheijden, and L. Yang, arXiv preprint arXiv:2402.03425 (2024)
2024
-
[10]
D. N. Page, Journal of Cosmology and Astroparticle Physics 2013, 028 (2013)
2013
-
[11]
Bianchi and P
E. Bianchi and P. Don` a, Phys. Rev. D 100, 105010 (2019)
2019
-
[12]
Blake and A
M. Blake and A. P. Thompson, Journal of High Energy Physics 2023, 1 (2023)
2023
-
[13]
Glatthard, Phys
J. Glatthard, Phys. Rev. D 109, L081901 (2024)
2024
-
[14]
Agarwal and N
K. Agarwal and N. Bao, Phys. Rev. D 102, 086017 (2020)
2020
-
[15]
Piroli, C
L. Piroli, C. S¨ underhauf, and X.-L. Qi, Journal of High Energy Physics 2020, 1 (2020)
2020
-
[16]
Kehrein, Phys
S. Kehrein, Phys. Rev. B 109, 224308 (2024)
2024
-
[17]
M. Saha, M. Kulkarni, and A. Dhar, Physical Review Letters 133, 230402 (2024)
2024
-
[18]
Glatthard, arXiv preprint arXiv:2501.09082 (2025)
J. Glatthard, arXiv preprint arXiv:2501.09082 (2025)
2025 arXiv
-
[19]
Ganguly, P
K. Ganguly, P. Gopalakrishnan, A. Naik, B. K. Agar- walla, and M. Kulkarni, arXiv preprint arXiv:2501.12110 (2025)
2025 arXiv
-
[20]
R. Jha, S. R. Manmana, and S. Kehrein, in preparation (2025)
2025
-
[21]
Li and F
H. Li and F. D. M. Haldane, Phys. Rev. Lett.101, 010504 (2008)
2008
-
[22]
Polkovnikov, K
A. Polkovnikov, K. Sengupta, A. Silva, and M. Vengalat- tore, Rev. Mod. Phys. 83, 863 (2011)
2011
-
[23]
J. R. Garrison and T. Grover, Phys. Rev. X 8, 021026 (2018)
2018
-
[24]
W. Zhu, Z. Huang, Y.-C. He, and X. Wen, Physical review letters 124, 100605 (2020)
2020
-
[25]
Kim and D
H. Kim and D. A. Huse, Physical review letters 111, 127205 (2013)
2013
-
[26]
X. Mi, A. Michailidis, S. Shabani, K. Miao, P. Klimov, J. Lloyd, E. Rosenberg, R. Acharya, I. Aleiner, T. An- dersen, et al., Science 383, 1332 (2024)
2024
-
[27]
Fukunaga, Introduction to statistical pattern recogni- tion (Elsevier, 2013)
K. Fukunaga, Introduction to statistical pattern recogni- tion (Elsevier, 2013)
2013
-
[28]
V. B. Bulchandani, S. Gopalakrishnan, and E. Ilievski, Journal of Statistical Mechanics: Theory and Experi- ment 2021, 084001 (2021)
2021
-
[29]
Elben, S
A. Elben, S. T. Flammia, H.-Y. Huang, R. Kueng, J. Preskill, B. Vermersch, and P. Zoller, Nature Reviews Physics 5, 9 (2023)
2023
-
[30]
M. K. Joshi, C. Kokail, R. van Bijnen, F. Kranzl, T. V. Zache, R. Blatt, C. F. Roos, and P. Zoller, Nature 624, 539 (2023)
2023
-
[31]
Kokail, R
C. Kokail, R. van Bijnen, A. Elben, B. Vermersch, and P. Zoller, Nature Physics 17, 936 (2021)
2021
-
[32]
Brydges, A
T. Brydges, A. Elben, P. Jurcevic, B. Vermersch, C. Maier, B. P. Lanyon, P. Zoller, R. Blatt, and C. F. Roos, Science 364, 260 (2019)
2019
-
[33]
M. K. Joshi, A. Elben, B. Vermersch, T. Brydges, C. Maier, P. Zoller, R. Blatt, and C. F. Roos, Phys. Rev. Lett. 124, 240505 (2020)
2020
-
[34]
Almheiri, N
A. Almheiri, N. Engelhardt, D. Marolf, and H. Maxfield, J. High Energy Phys. 12, 1 (2019). 8
2019
-
[35]
Almheiri, T
A. Almheiri, T. Hartman, J. Maldacena, E. Shaghoulian, and A. Tajdini, J. High Energy Phys. 5, 1 (2020)
2020
-
[36]
6yH12Vlz0niSDD67+EmJBnj08nE=
A. Almheiri, R. Mahajan, J. Maldacena, and Y. Zhao, J. High Energy Phys. 3, 1 (2020). Supplemental Materials: Sharp Page transitions in generic Hamiltonian dynamics Lauren H. Li, 1 Stefan Kehrein, 2 and Sarang Gopalakrishnan 3 1Department of Physics, Princeton University, Prin...
2020
Reviewed August 9, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.