{"id":"27319035-e484-4a61-94de-f2ecf85357b4","arxiv_id":"math-ph/0603006","paper_version":2,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"A rigorous proof that finite-dimensional quantum systems coupled to two bosonic reservoirs at different temperatures converge to a stable non-equilibrium stationary state with positive entropy production, when temperatures exceed a small power of the coupling.","lead":"This mathematics paper proves that a small quantum system coupled to two heat reservoirs at different temperatures settles into a steady non-equilibrium state, with heat flowing from the hotter to the colder reservoir and positive entropy production. It gives rigorous conditions for this behavior and shows the steady state is dynamically stable.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 3.2's entropy-positivity proof silently requires µ>3/2, while the theorem is stated under Theorem 3.1's weaker condition µ>1; the claimed leading-order expansion (12.3) is unsupported for 1<µ≤3/2.","rationale":"The Reader's weakest assumption, Condition (B), is real and is accurately described in the paper as a severe limitation on coupling functions; but it is an explicit, acknowledged hypothesis, and the theorems are conditional on it. The more load-bearing issue is internal: Theorem 3.2 states a conclusion under the hypotheses of Theorem 3.1 (µ>1), yet its proof depends on Theorem 12.1, whose proof restricts to µ>3/2. The text after Theorem 3.2 even says that for µ>3/2 the analysis gives the stronger leading-order computation, which supports the reading that the restrictive condition was not carried back into the theorem statement. Since strict positivity of entropy production is one of the two principal claims in the abstract, an unproven parameter range is a substantive correctness concern, not a cosmetic one. The paper's central existence and stability theorem (Theorem 3.1) does not appear to be affected, so this is a conditional-accept issue rather than a rejection: the manuscript should either add µ>3/2 to Theorem 3.2 or complete the missing estimate for 1<µ≤3/2. I also note the paper contains no machine-checked or reproducible computational verification, so the gap cannot be dismissed by external evidence; the proposed re-derivation is the decisive check.","tokens_in":49855,"tokens_out":9620,"duration_ms":81566,"concrete_test":"Re-derive the remainder estimate in Section 12 without the µ>3/2 restriction, e.g. for µ=4/3 (α=5/11). Take ρ0=g^{2−2α}=g^{12/11} and insert it into (12.8)–(12.12). If the remainder is O(g^{1+2α}δβ^2)=O(g^{21/11}δβ^2) rather than o(g^2)O(δβ), then (12.3) does not follow for this µ and Theorem 3.2 must be restated with the hypothesis µ>3/2 (α>1/2) or supplied with an additional argument. If the estimate is instead o(g^2) for all µ>1 after the telescoping step (12.22), the concern dissolves.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"Theorem 3.2 is stated under the conditions of Theorem 3.1, which require only µ>1; equivalently α=(µ−1/2)/(µ+1/2)>1/3. The proof of Theorem 3.2 at the end of Section 12, however, invokes Theorem 12.1, and the proof of Theorem 12.1 explicitly begins by restricting to µ>3/2, i.e. α>1/2. The restriction is not cosmetic. With ρ0=g^{2−2α}, the error terms from the expansions (11.7) and (11.15) are controlled by (gδβ ρ0^{−1/2})^2 = g^{2α} δβ^2; after multiplication by φ1 = O(g) this contributes g^{1+2α} δβ^2. Such a term is o(g^2) only when α>1/2. For 1<µ≤3/2 the paper gives no estimate showing η(φ1)=g^2 η′+o(g^2)O(δβ), Eq. (12.3), and the strict positivity conclusion EP(η)>0, which uses the sign of the leading coefficient η′, is therefore not established in that parameter range. The abstract advertises the entropy-production result for 0<α<1, but the proof as written supports it only for α>1/2. This is an internal gap in the stated Theorem 3.2, not merely a limitation of the model class.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies a finite-dimensional quantum system coupled to two bosonic reservoirs at inverse temperatures β1 and β2. In the Araki–Woods representation it constructs a non-self-adjoint Liouville operator K and a two-parameter complex deformation Kθ, then uses spectral information on Kθ, level shift operators, and a Feshbach-map expansion to characterize the non-equilibrium stationary state (NESS). Theorem 3.1 asserts that, under Conditions (A)–(E) and for 0<|g|<g1 with μ>1, every β1β2-normal initial state converges weakly to a stationary state η on a strongly dense subalgebra, with exponential convergence on a suitable dense set. Theorem 3.2 asserts that if β1≠β2, if g and |β1−β2| are sufficiently small independently, and if either G1=G2 or dimHp=2, then the entropy production EP(ηβ1β2) is strictly positive. The proof of the positivity statement proceeds through a leading-order expansion of the heat flux, η(φ1)=g²η′+o(g²)O(δβ), with an explicit formula for η′ in terms of the level-shift eigenvector ζ∗. The abstract advertises the result for 0<α<1, where α=(μ−1/2)/(μ+1/2).","tokens_in":50093,"tokens_out":7385,"duration_ms":67935,"significance":"If the results are valid, this is a substantial contribution to the rigorous theory of non-equilibrium quantum statistical mechanics. It extends the scattering-based construction of NESS for Fermi reservoirs of Jakšić–Pillet to bosonic reservoirs, improves the temperature lower bound from logarithmic in 1/g to a power g^{2+α}, and provides a quantitative leading-order expression for the heat current, Eq. (12.5), which is a valuable and falsifiable prediction. The paper is careful and honest about its limitations: finite-dimensional particle system, restriction to translation-analytic couplings, and high temperatures are all explicitly acknowledged. The central construction is non-circular: the NESS is obtained from a resonance eigenvector of a deformed Liouville operator and its properties are derived, not assumed. The main obstacle I find is not in the conceptual framework but in a parameter-range mismatch in the entropy-production proof, discussed in the major comments.","major_comments":[{"comment":"Theorem 12.1 is stated under the conditions of Theorem 3.1, which require only μ>1, i.e. α=(μ−1/2)/(μ+1/2)>1/3. However, the proof begins by restricting to μ>3/2, i.e. α>1/2, and later uses the estimate (gδβρ0^{−1/2})²=o(g) with ρ0=g^{2−2α}, which is valid only for α>1/2. For 1<μ≤3/2, the error term from the expansion (11.7) after multiplication by φ1=O(g) is of order g^{1+2α}δβ², which is not o(g²). Consequently the leading-order expansion (12.3) is not established in that range, and the strict positivity conclusion of Theorem 3.2, which relies on the sign of the leading coefficient η′, is unsupported for 1<μ≤3/2. This is an internal gap in the stated theorem, not merely a restriction of the model. The authors should either extend the proof to the full range μ>1 or revise Theorem 3.2, the abstract, and the introduction to state the result for μ>3/2 (α>1/2).","section":"Section 12, Theorem 12.1 and proof"},{"comment":"Even within the restricted range μ>3/2, the passage from the rough estimate (12.16) to the refined estimate (12.3) is only sketched, with the text saying that the proof is given 'without going into much detail'. Since this refinement is precisely what justifies positivity when δβ is small, and since Theorem 3.2 requires δβ and g to be small independently, the derivation of the error term in (12.3) should be written out sufficiently completely for the reader to verify that no term of order g^{1+2α}δβ² or worse survives. If such a term does survive, then the restriction to μ>3/2 is not cosmetic.","section":"Section 12, proof of Theorem 12.1"}],"minor_comments":[{"comment":"The symbol δβ is used with two different meanings: δβj=β−βj in Condition (B) and formula (3.4), while δβ=β1−β2 in Theorem 3.2 and Section 12. Please disambiguate these notations.","section":"Sections 3 and 12"},{"comment":"The parameter μ is used as a truncation index in the approximating creation operators a∗j,n(Gj), while the same letter μ denotes the infrared regularity exponent in Condition (B). Renaming one of these quantities would avoid confusion.","section":"Section 2, Eq. (2.14)"},{"comment":"In the case dimHp=2, the proof of Theorem 12.2 states that expanding e^{β1E21}−α(E21) around δβ=0 'verifies' the positivity of the linear term, but the computation is not shown. A short explicit formula would make the argument easier to check.","section":"Section 12, Theorem 12.2"},{"comment":"Theorem 6.1 and related spectral results are quoted from the companion paper [25] without proof. Since Theorems 3.1 and 3.2 rely on them, the manuscript should list precisely which statements are imported from [25] and verify that those statements cover the temperature regime used here.","section":"Section 6, Theorem 6.1"}],"recommendation":"major_revision","confidential_remarks":"The parameter-range mismatch in the entropy-production proof is the only substantive obstacle I see to accepting the paper. The existence and dynamical stability result (Theorem 3.1) appears well supported, and the paper is honest about its limitations. I would be willing to accept after the authors either prove (12.3) for 1<μ≤3/2 or explicitly restate Theorem 3.2 and the abstract for μ>3/2. The referee should also confirm that the spectral results imported from [25] are available in published form, since Theorem 3.1 depends on them."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things you should know. First, this is the real article: existence and dynamical stability of non-equilibrium stationary states for a finite system coupled to two bosonic reservoirs, with temperature bound min T > g^(2+α), genuinely better than the logarithmic bound in Jaksic-Pillet. The machinery is substantial: a two-parameter complex deformation (dilation plus translation), Feshbach analysis, level shift operators, and a deformation-analytic construction of the NESS vector. The paper is honest about its limitations: finite-dimensional particle system, translation analyticity used in one place, and high temperature. The central existence and stability theorem looks solid; I did not find a hidden assumption there.\n\nSecond, the entropy production theorem as stated is overbroad. Theorem 3.2 is asserted under the conditions of Theorem 3.1, which only require µ > 1. But the proof of Theorem 12.1, on which Theorem 3.2 relies, explicitly restricts to µ > 3/2 (equivalently α > 1/2). The error estimate uses (g δβ ρ0^{-1/2})^2 = o(g) with ρ0 = g^{2-2α}, and that only holds for α > 1/2. For 1 < µ ≤ 3/2 the expansion (12.3) is unsupported, so the strict positivity conclusion EP(η) > 0 is not established in that range. The abstract advertises all 0 < α < 1. This is an internal gap in the stated theorem, not merely a limitation of the model class. It looks fixable—either with a refined estimate or by restricting the statement—but as written the proof does not match the claim.\n\nWhat else? The heavy reliance on [25] (same authors) for spectral properties of Kθ is real, but [25] is a published JFA paper, so it is not a circularity problem. Condition (B), translation analyticity, is restrictive and the paper itself flags it clearly; the reader is right that this is the main technical assumption. The reader's overall ACCEPT verdict is reasonable, but with the µ > 3/2 gap made explicit, I would want the authors to address it before publication.\n\nWho is this for? Mathematical physicists working on rigorous non-equilibrium quantum statistical mechanics, especially those following the resonance/deformation approach. It deserves a serious referee: the main construction is original and careful, and the entropy production gap is a well-defined technical issue that a competent referee can evaluate. Send it to peer review, with instructions to the authors to either prove the expansion for 1 < µ ≤ 3/2 or restrict Theorem 3.2 and the abstract accordingly.","headline":"A rigorous and substantive NESS construction for bosonic reservoirs, but the entropy production theorem is stated more broadly than its proof supports: the key expansion is only shown for µ > 3/2, not the µ > 1 assumed in Theorem 3.2.","tokens_in":124,"tokens_out":2588,"would_cite":true,"duration_ms":322237,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"pith_extraction":{"msc":["81Q10","82C10","82B10"],"pacs":[],"model":"deepseek-v4-flash","headline":"A finite quantum system weakly coupled to two heat reservoirs at different temperatures has a unique, dynamically stable non-equilibrium stationary state with nonvanishing heat flux and strictly positive entropy production.","keywords":["non-equilibrium stationary state","NESS","Liouville operator","complex deformation","Feshbach map","entropy production","heat flux","bosonic reservoirs"],"falsifier":"Compute the spectrum of $K_\\theta$ for a coupling form factor that obeys the ultraviolet, non-degeneracy, and Fermi golden rule conditions but is only finitely differentiable (for example, a compactly supported cutoff), and check whether zero is still a simple isolated eigenvalue for $\\mathrm{Im}\\,\\theta>0$ with $\\tau'\\gg g^{2+\\alpha}$; if the zero resonance instead sticks to the continuum threshold, the proof of Theorem 3.1 fails.","tokens_in":49596,"feed_emoji":"♨️","tokens_out":10913,"duration_ms":93747,"temperature":0.7,"pith_summary":"This paper tries to establish a resonance theory of non-equilibrium stationary states: a small quantum system coupled weakly to two bosonic heat reservoirs at temperatures $T_1$ and $T_2$ has a unique stationary state $\\eta_{\\beta_1\\beta_2}$ that every normal initial state approaches at large times. The state is built from a zero resonance of a non-self-adjoint Liouville operator, and when $T_1\\neq T_2$ it carries a nonzero heat flux and strictly positive entropy production. The results are proved for weak coupling $0<|g|<g_1$ and reservoir temperatures bounded below by $\\min(T_1,T_2)>g^{2+\\alpha}$. A sympathetic reader would care because the proof identifies the NESS spectrally, as a resonance, rather than through a variational principle, and it extends prior Fermi-reservoir results to bosonic reservoirs with a weaker high-temperature restriction.","feed_headline":"Two-temperature bath forces a unique nonequilibrium steady state","feed_subtitle":"Weak coupling to reservoirs at different temperatures still reaches a stationary state with positive entropy production.","key_machinery":"The workhorse is a non-self-adjoint Liouville operator $K$ on the positive-temperature representation Hilbert space, together with its complex deformation $K_\\theta=U_\\theta K U_\\theta^{-1}$, where $\\theta=(\\delta,\\tau)$ combines a dilation and a translation. The Feshbach map — an isospectral reduction that replaces part of an operator by an effective operator on a small spectral subspace — reduces $K_\\theta$ near zero to a level shift operator $\\Lambda_0$, an $N\\times N$ matrix acting on the degenerate zero-eigenspace of the free generator. The zero eigenvector of $\\Lambda_0^*$ gives the NESS vector; the one-dimensional kernel and the spectral gap of $\\Lambda_0$ at zero control the isolation of the resonance, while the lower bounds on $\\mathrm{Im}\\,\\Lambda_e$ for nonzero $e$ control the rest of the spectrum and produce exponential convergence.","core_discovery":"The central claim is that the NESS is a resonance, not a normal state. Under conditions (A)–(E), the authors construct a non-self-adjoint Liouville generator $K$ on the positive-temperature representation Hilbert space and show, via a combined dilation-translation deformation $K_\\theta$, that the long-time limit of any $\\beta_1\\beta_2$-normal state is $\\eta(A)=\\langle\\Omega^*_\\theta,\\pi(A)_\\theta\\Omega_\\theta\\rangle$, where $\\Omega_\\theta$ and $\\Omega^*_\\theta$ are the zero eigenvectors of $K_\\theta$ and $K^*_\\theta$. Theorem 3.1 states that the convergence holds on a strongly dense algebra $A_1$, is uniform on a deformation norm, and is exponential for a dense set of initial states. Theorem 3.2 states that if $\\beta_1\\neq\\beta_2$ and either the two coupling functions coincide or the particle system has two levels, then the entropy production $\\mathrm{EP}(\\eta)=(\\beta_1-\\beta_2)\\eta(\\varphi_1)$ is strictly positive, with the heat flux computed to leading order as $g^2\\eta'$ and $\\eta'$ given explicitly in formula (12.5).","pith_inferences":["Extension — If translation analyticity is only a technical convenience, a coupling form factor with finite regularity but still satisfying the ultraviolet and infrared conditions should still produce an isolated zero resonance; checking the spectrum of $K_\\theta$ for finitely differentiable $G_j$ would test whether Condition (B) can be relaxed.","Extension — The explicit formula (12.5) and the analyticity of $\\eta'$ in the temperatures suggest that strict positivity of entropy production holds for generic pairs of coupling functions, not only for $G_1=G_2$ or $\\dim H_p=2$; the paper's almost-everywhere positivity remark points in the same direction.","Extension — The single Feshbach step is announced as the first of an iterated spectral renormalization group; iterating it should remove the $g^{2+\\alpha}$ temperature floor and yield NESS stability at arbitrarily low temperatures.","Extension — The same zero-resonance construction should extend to more than two reservoirs, giving a multi-terminal NESS whose entropy production is $\\sum_j\\beta_j\\eta(\\varphi_j)$ with heat flowing from each hotter to each colder reservoir."],"forward_implications":["Any $\\beta_1\\beta_2$-normal initial state converges to the same stationary state $\\eta_{\\beta_1\\beta_2}$; the convergence is uniform on the analytic-observable algebra $A_1$ and exponentially fast on a dense set.","When $\\beta_1\\neq\\beta_2$, the heat flux $\\eta(\\varphi_1)$ does not vanish, and entropy production $\\mathrm{EP}(\\eta)>0$; equivalently, energy flows from the hotter to the colder reservoir.","The NESS $\\eta$ is generally not a normal state of the full algebra, so it is defined only on a strongly dense subalgebra $A_1$; its expectation values are analytic in the coupling constant $g$.","At equal temperatures the same construction yields an equilibrium state with zero entropy production, as the paper states via Araki's perturbation theory of KMS states.","The results hold for reservoir temperatures $\\min(T_1,T_2)>g^{2+\\alpha}$, which improves the earlier high-temperature bound $\\min_j T_j\\ge c[\\ln(1/g)]^{-1}$ reported in the paper."],"supporting_citations":[{"why":"Supplies the NESS construction for Fermi reservoirs and the key formula linking long-time dynamics to the zero eigenprojection of the Liouville generator, which this paper adapts to Bose reservoirs.","marker":"[21]"},{"why":"Gives the spectral analysis of complex-deformed Liouville operators for coupled heat reservoirs, including the resolvent bounds and level shift formulæ used for $K_\\theta$.","marker":"[25]"},{"why":"Provides the single-reservoir return-to-equilibrium estimates, including dilation analyticity and level shift operators, that the two-reservoir proof extends.","marker":"[8]"},{"why":"Establishes the isospectrality of the Feshbach map used to isolate the zero resonance and reduce the deformed operator to a finite-dimensional level shift operator.","marker":"[7]"},{"why":"Proves that the entropy production of any NESS is nonnegative, which together with the leading-order computation identifies the sign of the heat flux.","marker":"[19]"},{"why":"Analyzes level shift operators for open quantum systems, giving the structure and spectral gap at zero used in the proof of Theorem 7.2.","marker":"[24]"},{"why":"Introduces the translation-type deformation and the positive-temperature representation that Condition (B) makes analytic, a component of the combined deformation $K_\\theta$.","marker":"[18]"}],"fun_headline_variants":["Two-temperature baths yield stationary state as resonance","Nonequilibrium steady states: resonances, not normal states","Quantum system in two baths: unique NESS with positive entropy","Stable nonequilibrium state from two reservoirs of different temps","Resonance framework explains non-equilibrium stationary states"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument needs the coupling functions to be analytic under complex shifts of the frequency variable, in a strip whose width is capped by the lower reservoir temperature; without that translation analyticity, the zero resonance sits at the threshold of the continuous spectrum and the resolvent estimate on which the convergence proof rests breaks down.","fun_headline_variants_meta":{"raw":{"variants":["Two-temperature baths yield stationary state as resonance","Nonequilibrium steady states: resonances, not normal states","Quantum system in two baths: unique NESS with positive entropy","Stable nonequilibrium state from two reservoirs of different temps","Resonance framework explains non-equilibrium stationary states"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000197,"raw_usage":{"total_tokens":1390,"prompt_tokens":998,"completion_tokens":392,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":614,"completion_tokens_details":{"reasoning_tokens":312}},"tokens_in":614,"tokens_out":392,"duration_ms":4647,"temperature":1.0,"reasoning_tokens":312,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-28T19:37:40.105894+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the spectrum of $K_\\theta$ for a coupling form factor that obeys the ultraviolet, non-degeneracy, and Fermi golden rule conditions but is only finitely differentiable (for example, a compactly supported cutoff), and check whether zero is still a simple isolated eigenvalue for $\\mathrm{Im}\\,\\theta>0$ with $\\tau'\\gg g^{2+\\alpha}$; if the zero resonance instead sticks to the continuum threshold, the proof of Theorem 3.1 fails.","supporting_citations":[{"cited_title":"Jakˇ si´ c and C.-A","cited_arxiv_id":null,"evidence_quote":"Supplies the NESS construction for Fermi reservoirs and the key formula linking long-time dynamics to the zero eigenprojection of the Liouville generator, which this paper adapts to Bose reservoirs."},{"cited_title":"Merkli, M","cited_arxiv_id":null,"evidence_quote":"Gives the spectral analysis of complex-deformed Liouville operators for coupled heat reservoirs, including the resolvent bounds and level shift formulæ used for $K_\\theta$."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the single-reservoir return-to-equilibrium estimates, including dilation analyticity and level shift operators, that the two-reservoir proof extends."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes the isospectrality of the Feshbach map used to isolate the zero resonance and reduce the deformed operator to a finite-dimensional level shift operator."},{"cited_title":"Jakˇ si´ c and C.-A","cited_arxiv_id":null,"evidence_quote":"Proves that the entropy production of any NESS is nonnegative, which together with the leading-order computation identifies the sign of the heat flux."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Analyzes level shift operators for open quantum systems, giving the structure and spectral gap at zero used in the proof of Theorem 7.2."},{"cited_title":"Jakˇ si´ c and C.-A","cited_arxiv_id":null,"evidence_quote":"Introduces the translation-type deformation and the positive-temperature representation that Condition (B) makes analytic, a component of the combined deformation $K_\\theta$."}],"review_version":1}