Pith. sign in

REVIEW 3 major objections 3 minor 110 references

A quantum channel's free energy—its relative entropy to a perfectly thermal channel—exactly sets its asymptotic athermality distillation and formation rates, making the resource theory reversible.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

T0 review · deepseek-v4-flash

2026-08-04 09:53 UTC pith:ALHXPZUA

load-bearing objection A serious framework for channel free energy with a fixable gap in the one-shot distillation proof; the additivity worry from the stress test is resolved by a valid chain-rule lemma. the 3 major comments →

arxiv 2510.12790 v4 pith:ALHXPZUA submitted 2025-10-14 quant-ph cond-mat.stat-mechhep-thmath-phmath.MP

Thermodynamics of quantum processes: An operational framework for free energy and reversible athermality

classification quant-ph cond-mat.stat-mechhep-thmath-phmath.MP
keywords quantum channelsathermalityfree energyGibbs-preserving superchannelsresource theoryasymptotic reversibilitywork extractionquantum relative entropy
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper extends free energy from quantum states to quantum processes (channels). It defines a channel's free energy as inverse temperature times its relative entropy to the absolutely thermal channel—the channel that thermalizes every input. The central claim is that this quantity is not merely formal: in a resource theory where Gibbs-preserving superchannels are free, the asymptotic rate at which a channel's athermality can be distilled into golden units equals the rate at which golden units must be spent to form the channel, and both equal half the relative entropy (equivalently β/2 times the channel free energy). A sympathetic reader should care because this makes athermality of processes a convertible, reversible resource in the asymptotic limit, and ties free energy directly to maximal work extraction via partial thermalization.

Core claim

On the paper's own terms, the discovery is Theorem 3: for any square quantum channel N and inverse temperature β, the nonadaptive and adaptive asymptotic athermality distillation rates and the nonadaptive formation rate all coincide: C^ε,∥_distill[N] = C^ε,ad_distill[N] = C^ε,∥_cost[N] = (1/2) D[N||T^β] = (β/2) F^β[N]. Theorem 5 adds that F^β[N] equals W^ext_{N→T^β}, the maximal work extractable by partially thermalizing the channel output. Thus the relative-entropy quantity defined in Eq. (3) is the operational free energy of a quantum process, and the resource theory of athermality under Gibbs-preserving superchannels is asymptotically reversible.

What carries the argument

The load-bearing objects are: the absolutely thermal channel T^β (a replacer channel that outputs the thermal state γ^β for every input), the channel relative entropy D[N||T^β] = sup_ψ D(id⊗N(ψ) || id⊗T^β(ψ)), the resource-theoretic free energy F^β[N] = β^{-1}D[N||T^β], and the golden unit (id_m, R^π)—the identity channel paired with the uniformly mixing (completely depolarizing) channel, with the output Hamiltonian chosen fully degenerate so this unit has minimal free energy among unitary channels. Free operations are Gibbs-preserving superchannels: superchannels that map T^β to itself. The machinery works because one-shot distillation and formation reduce to hypothesis-testing and max-rela

Load-bearing premise

The whole framework assumes that arbitrary Gibbs-preserving superchannels—including the measurement-and-reprepare operation used in the distillation protocol—are free, so a zero-cost engine can realize them; if physical thermal operations are the actual free operations, the identity channel has zero distillable work and the claimed reversibility no longer holds.

What would settle it

A concrete check: take the identity channel on a qubit and ask whether (id, R^π) can be formed from the free object T^β at positive rate. Under the paper's Gibbs-preserving-superchannel model the identity is maximally resourceful; under thermal operations the paper notes its distillable work is zero, so the same channel would carry two different 'free energies.' More quantitatively, evaluate whether lim_{n→∞} (1/n) D^ε_∞[id^{⊗n}||(T^β)^{⊗n}] equals D[id||T^β]; if the smoothed max-relative entropy rate differs, the one-shot formation formula would fail to converge to Theorem 3's rate.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

  • For any square channel, repeated parallel uses erase the gap between making and using athermality: no asymptotic loss in interconversion, so athermality behaves like a currency with a single exchange rate.
  • Single-shot athermality distillation is exactly (1/2)D^ε_H[N||T^β] and formation is exactly (1/2)D^ε_∞[N||T^β], giving exact finite-size formulas.
  • The channel free energy is the maximum work extractable by partially thermalizing the output: F^β[N] = W^ext_{N→T^β}.
  • In the β→0 limit or with a degenerate output Hamiltonian, the framework reduces to the dynamical resource theory of purity, and the private randomness capacity of a channel equals D[N||R^π].
  • The free energy satisfies monotonicity, faithfulness, continuity, additivity, and convexity under tensor products, so it behaves thermodynamically and is computable via convex optimization.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The reversibility is stated relative to all Gibbs-preserving superchannels; if physical implementations are restricted to thermal operations, the same relative-entropy object stops being the conversion rate (the identity channel becomes free), so the claimed universality is conditional on that idealization.
  • Because unitary channels are the maximally resourceful and the identity channel is the golden unit, the framework implicitly links thermodynamic value to quantum information transmission: channels that preserve maximal correlations are also thermodynamically most valuable.
  • The trade-off relation between one-shot cost and distillation suggests a finite-size second law for channels: beyond asymptotics, a channel's athermality cannot be simultaneously distilled and re-formed without a penalty that scales like (1/2) ln(1/(1−ε)).
  • A concrete test: compute the one-shot rates for small channels (e.g., qubit amplitude damping) via the provided semidefinite program for max-relative entropy and compare to the asymptotic formula; any discrepancy for some ε would pinpoint where the reversibility argument needs an additional assumption.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

3 major / 3 minor

Summary. The paper develops a dynamical resource theory of athermality for quantum channels. The free object is the absolutely thermal replacer channel T^β, the free operations are Gibbs-preserving superchannels (GPSCs), and the golden unit is the identity channel paired with the uniformly mixing channel (id_m, R_π). The channel free energy is defined as F^β[N] = β^{-1}D[N||T^β], and the authors prove single-shot distillation and formation formulas in terms of hypothesis-testing and max-relative entropies (Theorem 2), asymptotic reversibility with rate ½D[N||T^β] (Theorem 3), and a work-extraction interpretation (Theorem 5). They also connect the framework to private randomness, purity distillation, and entropy/energy relations. The central operational claim is that D[N||T^β] is not merely a formal analogue but the exact asymptotic conversion rate of channel athermality under GPSCs.

Significance. If the central claims hold, this is a substantial contribution: it provides a reversible resource theory for quantum channels under a well-defined (though idealized) class of free operations, giving channel-level analogues of the state resource theory of athermality. The paper is transparent about the main idealization—the use of all GPSCs as free operations—and it offers explicit protocol constructions, including the measurement-and-prepare GPSC used for distillation. The connections to channel capacities, private randomness, and work extraction give the framework genuine operational breadth. However, the current proof of the one-shot characterization is incomplete, and the bridge from resource-theoretic rates to the cited channel-discrimination theorems is stated rather than demonstrated. These gaps are fixable within the manuscript's scope, but they need to be addressed before the results can be regarded as fully established.

major comments (3)
  1. [Appendix A4, Theorem 2] The proof of Eqs. (A48)–(A49) establishes only the achievability direction. It constructs the specific GPSC Θ^Λ_ψ in Eq. (A50) and evaluates the optimization over m, Λ, ψ, giving the stated entropy expressions. It does not show that an arbitrary GPSC in the definitions (24)–(25) cannot achieve a larger distillation or a smaller formation cost. The equality chain in (A51)–(A55) implicitly assumes that the minimization over all Θ reduces to this measurement-and-prepare form. That reduction is plausible but requires a proof (e.g., via a one-shot channel-discrimination argument). Please supply the converse or a precise reduction argument.
  2. [Theorem 3, Eq. (33)] The proof consists of two citations, [65, Theorem 1] and [58, Theorem 4.1], with no explicit derivation of the correspondence between the resource-theoretic rates (29)–(32) and the channel-discrimination quantities in those papers. In particular, the factor 1/2, the difference between adaptive and parallel strategies, and the fact that arbitrary GPSCs are allowed in (24)–(25) need to be made explicit. This correspondence is the foundation of the asymptotic reversibility claim, so it should be stated as a formal lemma with a proof rather than assumed.
  3. [Appendix A1, additivity proof] The additivity property (A5) for α = 1 is load-bearing for the single-copy rate in Eq. (33). The proof uses the inequality D_α(N(ρ)||σ⊗τ) ≤ D_α[N||R_τ] + D_α(ρ||σ) attributed to [67]. This is a strong lemma; please state it precisely and confirm that its hypotheses are satisfied for α = 1, arbitrary channel N, and the replacer T^β. For the record, the concern that D[N||T^β] may be superadditive because it contains a coherent-information term does not land here, since Proposition 4 uses the mutual information I(R;A), not the coherent information. Nevertheless, the additivity proof would benefit from being self-contained enough to explicitly rule out superadditivity.
minor comments (3)
  1. [Lemma 2, Eq. (9)] The expression D∞(Φ^N_RA∥πA ⊗ γ^β_A) is likely a typo: the first factor in the second argument should be π_R, the maximally mixed state on the reference, not π_A. Please check the notation.
  2. [Section IV, Eqs. (29)–(33)] The notation for the asymptotic cost rates is inconsistent: Eq. (33) states C^{ε,∥}_distill = C^{ε,ad}_distill = C^{ε,∥}_cost = ½D[N||T^β], but no equality for C^{ε,ad}_cost is claimed. The authors should clarify whether the adaptive cost rate is also equal or whether it is merely bounded below by the other rates.
  3. [General presentation] There are several typographical issues: 'sandiwched' in Section VI, 'Distill' vs 'Dist' in Eqs. (29)–(32), and the repeated missing superscripts in some displayed equations (e.g., 'bγβ' in Definition 3). A careful proofreading pass is recommended.

Circularity Check

0 steps flagged

No load-bearing circularity; the central operational identifications rest on independently defined tasks and external channel-discrimination AEP results.

full rationale

The free energy F^β[N] := β^{-1}D[N||T^β] (Eq. 3) is introduced as a definition, not derived from the operational tasks, so no self-definitional reduction occurs. The one-shot distillation and formation results (Theorem 2, Eqs. (26)-(27)) are proved from the independently defined tasks in Eqs. (24)-(25) via hypothesis-testing and max-relative entropies; the proof explicitly constructs a GPSC but does not substitute F^β for the task. The asymptotic reversibility result (Theorem 3, Eq. (33)) is justified by citing external channel-discrimination AEP theorems ([65, Theorem 1] and [58, Theorem 4.1]), not by the paper's own additivity axiom or by the definition of F^β; therefore the equality of rates with D[N||T^β] is imported from independent results rather than being true by construction. The work-extraction identification (Theorem 5) follows from Proposition 4 together with standard state work-extraction formulas, and the partial-thermalization task is defined without reference to the channel free energy. The choice of Gibbs-preserving superchannels as free operations is an explicit modeling assumption, transparently acknowledged and contrasted with thermal operations; an idealization is not circularity. The skeptic's concern about possible non-additivity of D[N||T^β] and the distinction between single-copy and regularized divergences is a mathematical correctness question about the cited AEP theorems, not an instance of a prediction being forced by a fitted input or a self-citation chain. Some definitions are credited to the authors' prior work ([25], [30], [31]), but these citations supply starting quantities rather than the reversibility theorem itself, so they are not load-bearing circularity.

Axiom & Free-Parameter Ledger

0 free parameters · 6 axioms · 2 invented entities

The central mathematical object is a definition (relative entropy to T^β), not a fitted model: there are no free parameters fitted to data. β and bH_A are physical inputs; the golden-unit Hamiltonian c1_A is a convention. The load-bearing inputs the paper does not prove: (1) the choice of GPSCs as free operations (a modeling axiom on which reversibility depends); (2) the asymptotic channel-discrimination theorems of Cooney-Mosonyi-Wilde and Fawzi-Gao-Rahaman, cited for Theorem 3; (3) the decoupling-work identity β^{-1}I(A;R) and the quench/isothermal protocol used in Theorem 5; (4) the monotonicity/additivity properties of sandwiched Rényi divergences for Theorem 1. No exotic entities are postulated; T^β and the golden unit are mathematical reference constructs with precedent ([25],[31],[59]).

axioms (6)
  • domain assumption Free object is the absolutely thermal channel T^β and free operations are Gibbs-preserving superchannels (GPSCs)
    Central modeling choice (Section IV). Reversibility and golden-unit equivalence rely on this; with thermal operations instead, results differ sharply (refs [19,70]).
  • domain assumption Maximum extractable work from erasing correlations (decoupling) is β^{-1} I(A;R)_ρ (refs [84,85]) and the quench/isothermal protocol realizes the free-energy differences
    Used in Theorem 5's operational work extraction; standard results imported from the literature.
  • standard math Asymptotic channel discrimination theorem [65, Thm 1] and channel AEP [58, Thm 4.1] giving the reversible rates
    Theorem 3 is a direct citation of these external theorems (Eqs. (34)-(35)).
  • standard math Sandwiched Rényi divergences are monotone, additive, and (quasi-)convex as per [30,58,67,93,94]
    Used for Theorem 1 (axioms A1-A6) and Lemma 1.
  • domain assumption All systems finite-dimensional and the resource theory restricted to square channels (|A'| = |A|)
    Golden units and the Weyl decomposition require finite dimension m and square channels (Section IV).
  • domain assumption Reference R non-interacting with channel output A for thermodynamic interpretation (bH^int_RA = 0)
    Used in Theorem 4 and the work extraction protocol in Section VI; without it, E[N] and F^β_T relations change.
invented entities (2)
  • Absolutely thermal channel T^β (replacer outputting thermal state γ^β) independent evidence
    purpose: Free object of the resource theory; the reference point for channel free energy and athermality
    Channel-level analog of the thermal state, introduced in the authors' prior works [25,31]; operationally it is the channel that thermalizes any input. Not a new physical entity; a mathematical reference point with reduction to the thermal state for preparation channels.
  • Golden unit (id_m, R_π) — identity channel vs uniformly mixing channel no independent evidence
    purpose: Standard unit in which distillation and formation rates are measured
    A normalization choice; all unitary channels are equivalent to it under GPSCs. Its free energy (2β^{-1}ln m with degenerate Hamiltonian) is a convention-dependent reference value, not a falsifiable physical prediction.

pith-pipeline@v1.3.0-alltime-deepseek · 33059 in / 32734 out tokens · 265204 ms · 2026-08-04T09:53:17.380452+00:00 · methodology

0 comments
read the original abstract

We explore the thermodynamics of quantum processes (quantum channels) by axiomatically introducing the free energy for channels, defined via the quantum relative entropy with an absolutely thermal channel whose fixed output is in equilibrium with a thermal reservoir. This definition finds strong support through its operational interpretations in designated quantum information and thermodynamic tasks. We construct a resource theory of athermality for quantum processes, where free operations are Gibbs preserving superchannels and golden units are unitary channels with respect to absolutely thermal channel having fully degenerate output Hamiltonian. We exactly characterize the one-shot distillation and formation of quantum channels using hypothesis-testing and max-relative entropy with respect to the absolutely thermal channel. These rates converge asymptotically to the channel free energy (up to a multiplicative factor of half the inverse temperature), establishing its operational meaning and proving the asymptotic reversibility of the athermality. We show the direct relation between the resource theory of athermality and quantum information tasks such as private randomness and purity distillation, and thermodynamic tasks of erasure and work extraction. Our work connects the core thermodynamic concepts of free energy, energy, entropy, and maximal extractable work of quantum processes to their information processing capabilities.

Figures

Figures reproduced from arXiv: 2510.12790 by Dhanuja G.S., Himanshu Badhani, Siddhartha Das.

Figure 1
Figure 1. Figure 1: A bipartite system with a non-interacting Hamil [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

110 extracted references · 7 canonical work pages

  1. [1]

    channelized

    (16) The Pinsker’s inequality implies that, for any two quan- tum channels N , M ∈Ch(A′, A), D[N ∥M] ≥ 1 2 ∥N − M∥2 ⋄. (17) IV. DYNAMICAL RESOURCE THEOR Y OF A THERMALITY We now introduce framework and characterization of the thermodynamic resource theory of square quantum channels. A quantum channel NA′→A is a square quan- tum channel if |A′| = |A|; in o...

  2. [2]

    With respect to this new Hamiltonian, the state ρ is equilibrium state

    Quenching: We instantaneously change the Hamil- tonian from bH to −β−1 ln ρ. With respect to this new Hamiltonian, the state ρ is equilibrium state. The work done on the system during the quench is the change in the energy of the system, given by δWquech = δ tr(ρ bH) (75) = tr(ρ(−β−1 ln ρ − bH)) (76) = −β−1 tr(ρ ln ρ − ρ ln γβ) + β−1 ln Z β (77) = −F β(ρ)...

  3. [3]

    Since the state at the initial point of this process is at equilibrium, the state at the final point of the reversible process will be γβ

    Reversible, isothermal process : We quasistatically change the Hamiltonian from −β−1 ln Z β back to bH. Since the state at the initial point of this process is at equilibrium, the state at the final point of the reversible process will be γβ. The work done during this reversible process is given by δWrev = δF eq T = F eq T (γβ) − F eq T (ρ) = F eq T (γβ)....

  4. [4]

    In the task of decoupling, the information stored as correlation gets erased and only local in- formation content on reduced states remain

    Decoupling: The state ρRA is first decoupled to ρR ⊗ ρA. In the task of decoupling, the information stored as correlation gets erased and only local in- formation content on reduced states remain. This task of erasure releases heat [83], which can be used to extract useful work. The maximum extractable work from the decoupling of a bipartite state ρAB in ...

  5. [5]

    Partial quenching : We instantaneously change the Hamiltonian bHA to −β−1 ln ρA from which we can extract the work, W ext quench = F β T (ρA). (85)

  6. [6]

    The extractable work during this process is given by W ext rev = −F eq T (γβ A)

    We perform reversible, isothermal driving of system A from the Hamiltonian −β−1 ln ρA back to bHA. The extractable work during this process is given by W ext rev = −F eq T (γβ A). (86) The total extractable work during the partial thermal- ization process, id R ⊗N (ψRA′) → idR ⊗T β(ψRA′), is W ext N (ψRA′ )→T β (ψRA′ ) = β−1I(A; R)N (ψRA′ ) + F β(N (ψA′))...

  7. [7]

    X x pxN x∥T β # = sup ψRA′ Dα X x pxN x(ψ)∥T β(ψ) ! (A31) ≤ sup ψRA′ X x pxDα(N x(ψ)∥T β(ψ)). (A32) Using the definition of free energy of channels, we have F β α

    Proof of Theorem 1 We begin the proof by proving two useful lemmas em- ployed to prove the theorem. Lemma 8. Given two quantum channels NA′→A and MA′→A such that 1 2 ∥N − M∥⋄ ≤ ε, we have F β[N ] − F β[M] ≤ β−1(εK + h2(ε)), (A1) where ε ∈ [0, 1] and h2(ε) = −ε ln(ε) − (1 − ε) ln(1 − ε) is the binary Shannon entropy. K is a positive real number such that K...

  8. [8]

    The sandwiched R´ enyi free energy F β α [N ], α ∈ [ 1 2 , ∞), is nonincreasing under the action of Gibbs- subpreserving superchannel

    Proof of Lemma 1 Lemma. The sandwiched R´ enyi free energy F β α [N ], α ∈ [ 1 2 , ∞), is nonincreasing under the action of Gibbs- subpreserving superchannel. It remains invariant under the action of Gibbs preserving unitary superchannels. Proof. Given the channel NA′→A and absolutely thermal channel T β A′→A, the sandwiched R´ enyi relative entropy of a ...

  9. [9]

    Proof of Proposition 2 Lemma 10. Given a state ρ = P i pi |ψi⟩ ⟨ψi| where ψi are pure states, the max-relative entropy is bounded as follows D∞(ρ∥σ) ≥ ln max i pi ⟨ψi| σ−1 |ψi⟩ , (A38) D∞(ρ∥σ) ≤ ln X i pi ⟨ψi| σ−1 |ψi⟩ ! . (A39) Proof. The max-relative entropyD∞(ρ∥σ) can be written as D∞(ρ∥σ) = ln ∥ X i piσ− 1 2 |ψi⟩ ⟨ψi| σ− 1 2 ∥∞ (A40) = ln ∥AA†∥∞ (A41)...

  10. [10]

    Proof of Theorem 2 Theorem. For any error ε ∈ [0, 1] and a given resource channel (N , T β), the single-shot athermality distillation and formation are proportional to the ε-hypothesis-testing free energy and the ε-max-free energy of the channel N , respectively, Distε(N , T β) = 1 2 Dε H [N ∥Tβ], (A48) Costε(N , T β) = 1 2 Dε ∞[N ∥Tβ], (A49) F β,ε H [N ]...

  11. [11]

    Dual :    maximize tr(Φ N RAXRA) subject to tr(( πA ⊗ γβ A)XRA) ≤ 1, XRA ≥ 0

    SDP for the max-free energy The max-free energy F β ∞[N ] of a quantum channel NA′→A is half of the logarithm of the optimal value of the following semidefinite program (SDP) (strong duality): Primal :    minimize λ ∈ R subject to Φ N RA ≤ λ(πR ⊗ γβ A), λ ≥ 0. Dual :    maximize tr(Φ N RAXRA) subject to tr(( πA ⊗ γβ A)XRA) ≤ 1, XRA ≥ 0. ...

  12. [12]

    The max-free energy F β ∞[U ] of a unitary quan- tum channel UA′→A is F β ∞[U ] = β−1 ln tr h (γβ A) −1i

    Proof of Lemma 3 Lemma. The max-free energy F β ∞[U ] of a unitary quan- tum channel UA′→A is F β ∞[U ] = β−1 ln tr h (γβ A) −1i . (A63) If the Hamiltonian of A is trivial or in general bHA = c1 A for c ∈ R , then F β ∞[U ] = 2β−1 ln |A|. Proof. Using the property of the max-relative entropy of quantum channels βF β ∞[U ] = D∞[U ∥Tβ] = D∞(ΦU ∥ΦT β ), (A64...

  13. [13]

    Proof of Proposition 3 Proposition. For a quantum channel N and ε ∈ (0, 1) the athermality distillation and formation of a resource channel (N , T β) satisfy the following trade-off relation, Cost √ε(N , T β) ≤ Dist1−ε(N , T β) + 1 2 ln 1 1 − ε . (A67) Proof. For ε ∈ (0, 1) and ρ ∈ St(A), σ ∈ Pos(A), we have [58, 96] D √ε ∞ (ρ∥σ) ≤ D1−ε H (ρ∥σ) + ln 1 1 −...

  14. [14]

    Given a unitary channel UA′→A, the mini- mum value of its max-free energy F β ∞[U ] is achieved for the output with trivial Hamiltonian, bHA = c1 A

    F ree energy of a golden unit Lemma 11. Given a unitary channel UA′→A, the mini- mum value of its max-free energy F β ∞[U ] is achieved for the output with trivial Hamiltonian, bHA = c1 A. In par- ticular, min bHA F β ∞[U ] = 2β−1 ln |A|. (A75) Proof. From Lemma 3 we have βF β ∞[U ] = ln tr(γ−1 β ) (A76) = ln tr eβ bHA + ln tr e−β bHA . (A77) We analyze t...

  15. [15]

    T. D. Ladd, F. Jelezko, R. Laflamme, Y. Nakamura, C. Monroe, and J. L. O’Brien, Quantum computers, Na- ture 464, 45–53 (2010)

  16. [16]

    Das, Bipartite quantum interactions: Entan- gling and information processing abilities (2019), arXiv:1901.05895

    S. Das, Bipartite quantum interactions: Entan- gling and information processing abilities (2019), arXiv:1901.05895

  17. [17]

    S. Das, S. B¨ auml, M. Winczewski, and K. Horodecki, Universal limitations on quantum key distribution over a network, Physical Review X 11, 041016 (2021)

  18. [18]

    Proctor, K

    T. Proctor, K. Rudinger, K. Young, E. Nielsen, and R. Blume-Kohout, Measuring the capabilities of quan- tum computers, Nature Physics 18, 75 (2022)

  19. [19]

    Sadhu, M

    A. Sadhu, M. A. Somayajula, K. Horodecki, and S. Das, Practical limitations on robustness and scalability of quantum internet (2024), arXiv:2308.12739 [quant-ph]

  20. [20]

    Ho, Universal thermodynamics of degenerate quan- tum gases in the unitarity limit, Physical Review Letters 92, 090402 (2004)

    T.-L. Ho, Universal thermodynamics of degenerate quan- tum gases in the unitarity limit, Physical Review Letters 92, 090402 (2004)

  21. [21]

    J. P. Pekola, Towards quantum thermodynamics in elec- tronic circuits, Nature Physics 11, 118 (2015)

  22. [22]

    Chu and J

    Y. Chu and J. Cai, Thermodynamic principle for quan- tum metrology, Physical Review Letters 128, 200501 (2022)

  23. [23]

    Singh, and G

    Swati, U. Singh, and G. Chiribella, A resource theory of activity for quantum thermodynamics in the absence of heat baths (2023), arXiv:2304.08926 [quant-ph]

  24. [24]

    Singh, S

    U. Singh, S. Das, and N. J. Cerf, Partial order on passive states and Hoffman majorization in quantum thermody- namics, Physical Review Research 3, 033091 (2021)

  25. [25]

    Horodecki, M

    K. Horodecki, M. Winczewski, L. Sikorski, P. Mazurek, M. Czechlewski, and R. Yehia, Quantification of the en- ergy consumption of entanglement distribution (2025), arXiv:2507.23108 [quant-ph]

  26. [26]

    F. G. S. L. Brand˜ ao, M. Horodecki, N. H. Y. Ng, J. Op- penheim, and S. Wehner, The second laws of quantum thermodynamics, Proceedings of the National Academy of Sciences 112, 3275 (2015)

  27. [27]

    Lostaglio, K

    M. Lostaglio, K. Korzekwa, D. Jennings, and T. Rudolph, Quantum coherence, time-translation symmetry, and thermodynamics, Physical Review X 5, 021001 (2015)

  28. [28]

    Shiraishi and T

    N. Shiraishi and T. Sagawa, Quantum thermodynam- ics of correlated-catalytic state conversion at small scale, Physical Review Letters 126, 150502 (2021)

  29. [29]

    Moroder, O

    M. Moroder, O. Culhane, K. Zawadzki, and J. Goold, Thermodynamics of the quantum Mpemba effect, Phys- ical Review Letters 133, 140404 (2024)

  30. [30]

    Shiraishi and R

    N. Shiraishi and R. Takagi, Recovery of the second law in fully quantum thermodynamics (2025), arXiv:2510.05642 [quant-ph]

  31. [31]

    Sagawa and M

    T. Sagawa and M. Ueda, Second law of thermodynamics with discrete quantum feedback control, Physical Review Letters 100, 080403 (2008). 21

  32. [32]

    Campisi, P

    M. Campisi, P. Talkner, and P. H¨ anggi, Fluctuation the- orem for arbitrary open quantum systems, Physical Re- view Letters 102, 210401 (2009)

  33. [33]

    Navascu´ es and L

    M. Navascu´ es and L. P. Garc ´ ıa-Pintos, Nonthermal quan- tum channels as a thermodynamical resource, Physical Review Letters 115, 010405 (2015)

  34. [34]

    Uzdin, A

    R. Uzdin, A. Levy, and R. Kosloff, Equivalence of quan- tum heat machines, and quantum-thermodynamic signa- tures, Physical Review X 5, 031044 (2015)

  35. [35]

    F. G. S. L. Brand˜ ao, M. Horodecki, J. Oppenheim, J. M. Renes, and R. W. Spekkens, Resource theory of quan- tum states out of thermal equilibrium, Physical Review Letters 111, 250404 (2013)

  36. [36]

    Deffner and S

    S. Deffner and S. Campbell, Quantum thermodynamics: An introduction to the thermodynamics of quantum in- formation (2019), arXiv:1907.01596 [quant-ph]

  37. [37]

    Gour, Resources of the quantum world (2024), arXiv:2402.05474 [quant-ph]

    G. Gour, Resources of the quantum world (2024), arXiv:2402.05474 [quant-ph]

  38. [38]

    Gour and M

    G. Gour and M. M. Wilde, Entropy of a quantum chan- nel, Physical Review Research 3, 023096 (2021)

  39. [39]

    Pandey, U

    Sohail, V. Pandey, U. Singh, and S. Das, Funda- mental limitations on the recoverability of quantum processes, Annales Henri Poincar´ e 10.1007/s00023-025- 01590-y (2025), arXiv:2403.12947

  40. [40]

    Das and M

    S. Das and M. M. Wilde, Quantum reading capacity: General definition and bounds, IEEE Transactions on In- formation Theory 65, 7566–7583 (2019)

  41. [41]

    E. Kaur, S. Das, M. M. Wilde, and A. Winter, Extendibil- ity limits the performance of quantum processors, Phys- ical Review Letters 123, 070502 (2019)

  42. [42]

    K. Fang, O. Fawzi, R. Renner, and D. Sutter, Chain rule for the quantum relative entropy, Physical Review Let- ters 124, 100501 (2020)

  43. [43]

    Y. Li, J. Xing, D. Qu, H. Gao, L. Xiao, J.-M. Liu, Y. Xiao, and P. Xue, Temporal asymmetry in entangle- ment distillation, Physical Review Letters 10.1103/glc7- xy8t (2025)

  44. [44]

    S. Das, K. Goswami, and V. Pandey, Conditional en- tropy and information of quantum processes (2024), arXiv:2410.01740

  45. [45]

    Das and U

    S. Das and U. Sen, Maximum entropy principle for quan- tum processes (2025), arXiv:2506.24079 [quant-ph]

  46. [46]

    M. N. Bera, A. Riera, M. Lewenstein, Z. B. Khanian, and A. Winter, Thermodynamics as a consequence of infor- mation conservation, Quantum 3, 121 (2019)

  47. [47]

    Skrzypczyk, A

    P. Skrzypczyk, A. J. Short, and S. Popescu, Work extrac- tion and thermodynamics for individual quantum sys- tems, Nature Communications 5, 4185 (2014)

  48. [48]

    G. Gour, M. P. M¨ uller, V. Narasimhachar, R. W. Spekkens, and N. Yunger Halpern, The resource the- ory of informational nonequilibrium in thermodynamics, Physics Reports 583, 1–58 (2015)

  49. [49]

    D. Yang, K. Horodecki, and A. Winter, Distributed pri- vate randomness distillation, Physical Review Letters 123, 170501 (2019)

  50. [50]

    X. Yuan, P. Zeng, M. Gao, and Q. Zhao, One-shot dy- namical resource theory (2020), arXiv:2012.02781 [quant- ph]

  51. [51]

    Maillet, P

    O. Maillet, P. A. Erdman, V. Cavina, B. Bhandari, E. T. Mannila, J. T. Peltonen, A. Mari, F. Taddei, C. Jarzyn- ski, V. Giovannetti, and J. P. Pekola, Optimal probabilis- tic work extraction beyond the free energy difference with a single-electron device, Physical Review Letters 122, 150604 (2019)

  52. [52]

    Badhani, D

    H. Badhani, D. G. S, S. Choudhary, V. Anand, and S. Das, Erasure cost of a quantum process: A thermo- dynamic meaning of the dynamical min-entropy (2025), arXiv:2506.05307 [quant-ph]

  53. [53]

    Hiai and D

    F. Hiai and D. Petz, Entropy densities for Gibbs states of quantum spin systems, Re- views in Mathematical Physics 05, 693 (1993), https://doi.org/10.1142/S0129055X93000218

  54. [54]

    Gerisch, Internal symmetries and limiting Gibbs states in quantum lattice mean-field theories, Physica A: Sta- tistical Mechanics and its Applications 197, 284 (1993)

    T. Gerisch, Internal symmetries and limiting Gibbs states in quantum lattice mean-field theories, Physica A: Sta- tistical Mechanics and its Applications 197, 284 (1993)

  55. [55]

    Tasaki, Jarzynski relations for quantum systems and some applications (2000), arXiv:cond-mat/0009244 [cond-mat.stat-mech]

    H. Tasaki, Jarzynski relations for quantum systems and some applications (2000), arXiv:cond-mat/0009244 [cond-mat.stat-mech]

  56. [56]

    Morningstar, D

    A. Morningstar, D. A. Huse, and V. Khemani, Univer- sality classes of thermalization for mesoscopic Floquet systems, Physical Review B 108, 174303 (2023)

  57. [57]

    Bergamaschi and C.-F

    T. Bergamaschi and C.-F. Chen, Quantum spin chains thermalize at all temperatures (2025), arXiv:2510.08533 [quant-ph]

  58. [58]

    Goldstein, J

    S. Goldstein, J. L. Lebowitz, R. Tumulka, and N. Zangh ` ı, Canonical typicality, Physical Review Letters 96, 050403 (2006)

  59. [59]

    Linden, S

    N. Linden, S. Popescu, A. J. Short, and A. Winter, Quan- tum mechanical evolution towards thermal equilibrium, Physical Review E 79, 061103 (2009)

  60. [60]

    Hallam, J

    A. Hallam, J. G. Morley, and A. G. Green, The Lyapunov spectra of quantum thermalisation, Nature Communica- tions 10, 2708 (2019)

  61. [61]

    Badhani, D

    H. Badhani, D. G. S, and S. Das, Thermodynamic work capacity of quantum information processing (2025), arXiv:2510.23731 [quant-ph]

  62. [62]

    Chiribella, G

    G. Chiribella, G. M. D’Ariano, and P. Perinotti, Theoret- ical framework for quantum networks, Physical Review A 80, 022339 (2009)

  63. [63]

    Leditzky, E

    F. Leditzky, E. Kaur, N. Datta, and M. M. Wilde, Ap- proaches for approximate additivity of the Holevo in- formation of quantum channels, Physical Review A 97, 012332 (2018)

  64. [64]

    Tomamichel, Quantum information processing with finite resources – mathematical foundations (2021), arXiv:1504.00233v5

    M. Tomamichel, Quantum information processing with finite resources – mathematical foundations (2021), arXiv:1504.00233v5

  65. [65]

    M. M. Wilde, A. Winter, and D. Yang, Strong converse for the classical capacity of entanglement-breaking and Hadamard channels via a sandwiched R´ enyi relative en- tropy, Communications in Mathematical Physics 331, 593 (2014), arXiv:1306.1586

  66. [66]

    Datta, Min- and max-relative entropies and a new entanglement monotone, IEEE Transactions on Informa- tion Theory 55, 2816 (2009)

    N. Datta, Min- and max-relative entropies and a new entanglement monotone, IEEE Transactions on Informa- tion Theory 55, 2816 (2009)

  67. [67]

    Wang and R

    L. Wang and R. Renner, One-shot classical-quantum ca- pacity and hypothesis testing, Physical Review Letters 108, 200501 (2012)

  68. [68]

    Lenard, Thermodynamical proof of the Gibbs formula for elementary quantum systems, Journal of Statistical Physics 19, 575 (1978)

    A. Lenard, Thermodynamical proof of the Gibbs formula for elementary quantum systems, Journal of Statistical Physics 19, 575 (1978)

  69. [69]

    E. T. Jaynes, Information theory and statistical mechan- ics, Physical Review 106, 620 (1957)

  70. [70]

    R. K. Pathria and P. D. Beale, Statistical Mechanics: In- ternational Series of Monographs in Natural Philosophy (Elsevier, 2021)

  71. [71]

    Devetak, M

    I. Devetak, M. Junge, C. King, and M. B. Ruskai, Mul- tiplicativity of completely bounded p-norms implies a new additivity result, Communications in Mathematical 22 Physics 266, 37–63 (2006)

  72. [72]

    Fawzi, L

    O. Fawzi, L. Gao, and M. Rahaman, Asymptotic equipar- tition theorems in von Neumann algebras, Annales Henri Poincar´ e 10.1007/s00023-025-01545-3 (2025)

  73. [73]

    Y. Luo, S. Milz, and F. C. Binder, Thermodynamic criteria for signaling in quantum channels (2025), arXiv:2506.20428 [quant-ph]

  74. [74]

    Liu and A

    Z.-W. Liu and A. Winter, Resource theories of quantum channels and the universal role of resource erasure (2019), arXiv:1904.04201 [quant-ph]

  75. [75]

    Schumacher, Sending entanglement through noisy quantum channels, Physical Review A 54, 2614 (1996)

    B. Schumacher, Sending entanglement through noisy quantum channels, Physical Review A 54, 2614 (1996)

  76. [76]

    Abeyesinghe, I

    A. Abeyesinghe, I. Devetak, P. Hayden, and A. Winter, The mother of all protocols: restructuring quantum in- formation’s family tree, Proceedings of the Royal Soci- ety A: Mathematical, Physical and Engineering Sciences 465, 2537–2563 (2009)

  77. [77]

    Devetak and A

    I. Devetak and A. Winter, Distillation of secret key and entanglement from quantum states, Proceedings of the Royal Society A: Mathematical, Physical and Engineer- ing Sciences 461, 207–235 (2005)

  78. [78]

    S. Das, S. Khatri, G. Siopsis, and M. M. Wilde, Fun- damental limits on quantum dynamics based on en- tropy change, Journal of Mathematical Physics 59, 10.1063/1.4997044 (2018)

  79. [79]

    Cooney, M

    T. Cooney, M. Mosonyi, and M. M. Wilde, Strong converse exponents for a quantum channel discrimina- tion problem and quantum-feedback-assisted communi- cation, Communications in Mathematical Physics 344, 797 (2016), arXiv:1408.3373

  80. [80]

    Das and M

    S. Das and M. M. Wilde, Quantum rebound capacity, Physical Review A 100, 030302 (2019)

Showing first 80 references.