{"id":"0448227f-722c-4831-9abd-37500420637a","arxiv_id":"1908.06312","paper_version":1,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"For a nonreciprocal Kane-Lubensky chain, the zero-mode skin effect is predicted to coincide with a phase where the Petermann factor diverges exponentially with system size.","lead":"The paper develops a response theory for non-Hermitian mechanical metamaterials and predicts that the zero-mode skin effect is accompanied by a phase of extreme sensitivity to perturbations, with response growing without bound as the system gets large. This gives a new way to observe the system's left eigenvectors and points to sensing applications that avoid exceptional-point fine-tuning.","discovery_kind":"first_principles","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Eq. (12)'s finite-width regularization is load-bearing: the zero mode has no velocity damping at ω=0, so the η-shift creates the quasi-stationary response; the observable must be defined at fixed nonzero frequency, not via η.","rationale":"The reader's weakest assumption correctly identifies the finite-width regularization as the fragile step in translating K0 into an observable response, and I agree that this is the main caveat. The mathematical core is internally consistent: Eq. (15) follows from the biorthogonal zero modes, the Petermann factor is invariant under normalization choices, and the exponential divergence of K0 in the skin phase is independent of η. The issue is operational rather than algebraic: the physical zero mode is undamped at exactly zero frequency, so the quasi-stationary response used in Eq. (12) is not the response to a monochromatic zero-frequency drive. A direct test at fixed nonzero frequency, which does not need η, can settle whether the predicted divergence is observable; therefore the appropriate verdict is acceptance conditional on specifying and validating such a protocol. I am not recommending rejection because the divergence itself is real and the finite-frequency route is likely to confirm it. The reader's ACCEPT is reasonable, but the shift to CONDITIONAL makes the required operational clarification explicit without overstating the risk.","tokens_in":7704,"tokens_out":27111,"duration_ms":279584,"concrete_test":"Simulate or solve Eq. (5) directly, without any η-shift, for a=1, b=0.73, γ=0.02, at fixed drive frequencies ω=0.01 and ω=0.001, for N=9, 18, 36 and ε=0.1 (outside the skin phase) and ε=0.2 (inside it). Compute the stationary total power summed over output sites for a drive placed at the edge where the left zero-mode component is large, and compare its N-scaling with K0(ε,N)/[ω²(ω²+γ²)] at the same parameters. If the N-dependence follows K0 without invoking η, the concern is resolved; if the enhancement only appears when ω→0 first or when η is used, then the advertised response phase transition is not a directly measurable response property.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim is that the zero-mode skin phase produces a diverging physical response, quantified by the Petermann factor K0 of Eqs. (15) and (16). The translation of K0 into an observable response goes through Eq. (12), where the zero mode is regularized by a finite imaginary frequency shift η. In the physical equation of motion (5), the zero-mode amplitude obeys q̈ + γq̇ = F e^{-iωt}; at ω=0 the velocity-damping term vanishes and the zero mode has no stationary response (q grows linearly). Thus a monochromatic zero-frequency drive does not produce the quasi-stationary response implied by Eq. (12); the η-shift effectively supplies fictitious damping to exactly the mode that carries the predicted divergence. This is not merely cosmetic: at any finite nonzero ω the zero-mode denominator is ω²(ω²+γ²), so the response already diverges as ω→0 for all ε, and the skin-effect phase transition is cleanly defined only through K0, not through a directly measured displacement at a single frequency. The paper should specify an operational observable, such as a fixed small nonzero drive frequency or a finite measurement-time protocol, under which the measured N-scaling is K0/[ω²(ω²+γ²)].","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops a response theory for non-Hermitian nonreciprocal mechanical metamaterials described by a general dynamical matrix M. The central object is the regularized Green's function and the total power spectrum, which near a resonance is weighted by the biorthogonal Petermann factor K_k = (U^†U)_kk (U^{-1}U^{-†})_kk, where U contains the right eigenvectors and U^{-1} the left eigenvectors. Applied to the robotic metamaterial of Ghatak et al., the zero-mode profiles u0,n ∝ (a(1−ε)/b(1+ε))^n and v0,n ∝ (a/b)^n are used to derive Eq. (15) for K0, and Eq. (16) shows that K0 diverges exponentially with system size in the zero-mode skin-effect phase ε1 < ε < ε2, while remaining finite outside it. The paper thus predicts an extended phase of extreme low-frequency sensitivity that is tied to the non-Hermitian skin effect and not to exceptional points.","tokens_in":7950,"tokens_out":18082,"duration_ms":168504,"significance":"The result is significant because it identifies an experimentally accessible signature of biorthogonality in non-Hermitian metamaterials: the left eigenvectors and the Petermann factor, which are not directly visible in the right-eigenvector spatial profiles measured in the experiment, become manifest in the response to external driving. The core derivation is algebraic and parameter-free once the experimental couplings (a, b, ε) are fixed; Eq. (15) follows directly from biorthogonal normalization and the explicit mode profiles, and the divergence in the skin-effect phase is a falsifiable system-size-scaling prediction. The paper also correctly emphasizes that the sensitive phase is not tied to spectral exceptional points, which broadens the potential relevance to sensing applications. The main caveat is operational: because the zero mode has no velocity-dependent damping at exactly zero frequency, the statement of divergence should be phrased in terms of the system-size scaling of K0 at a fixed small nonzero drive frequency; with that interpretation, the central claim is sound.","major_comments":[],"minor_comments":[{"comment":"Equation (3): the right zero-mode profile should read u0,n = c_R [a(1−ε)/b(1+ε)]^n; the printed denominator b(1−ε) is inconsistent with Eq. (15) and with the stated phase boundaries ε1 = (a−b)/(a+b), ε2 = (a+b)/(a−b). Please correct this typo.","section":"Eq. (3)"},{"comment":"Paragraph following Eq. (5) and the discussion of Eq. (12): please add an explicit operational statement for the zero-mode response. At ω = 0 the stationary response is not defined because the velocity-dependent damping does not affect the zero mode; Eq. (13) shows that at fixed ω ≠ 0 the zero-mode contribution is P_tot ≈ K0/[ω²(ω²+γ²)], so the predicted phase transition is the system-size scaling of K0 at fixed small ω (or, equivalently, of the η-regularized response at ω = 0 with η fixed). A sentence stating this would remove the ambiguity about whether the divergence is an artifact of the η-regularization.","section":"Sec. 3 (around Eqs. (5)–(13))"},{"comment":"Figures 1(c) and 3: consider using a logarithmic color or vertical scale for K0, since the exponential growth across the skin-effect phase spans many orders of magnitude and the linear scale obscures the comparison with Eq. (16).","section":"Figs. 1(c) and 3"},{"comment":"In the introduction, 'compassing' should be 'encompassing'.","section":"Introduction"}],"recommendation":"minor_revision","confidential_remarks":"To the editor: This is a compact theory letter with a sound central derivation. The only substantive caveat is the operational definition of the zero-frequency response; I do not consider it fatal because Eq. (13) already provides a finite-frequency observable, but the authors should state it explicitly. The paper is within scope for a journal interested in non-Hermitian physics and metamaterials, and I have no concerns about novelty or citation practice."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Henning,\n\nThis paper is solid and worth engaging with. The main new thing is the observation that in the nonreciprocal mechanical metamaterial of Ghatak et al., the zero-mode skin effect comes with a large Petermann factor that diverges exponentially with system size across the whole skin-effect phase. The paper works out the response theory that separates left- and right-eigenvector contributions to the power spectrum, and shows that the total response is governed by the biorthogonal product K0. That is a clean, falsifiable prediction for the existing robotic metamaterial, and it does not rely on exceptional-point fine-tuning, which is a useful distinction.\n\nThe derivation is mostly algebra and is internally consistent. The mode profiles in Eq. (3) have a typo: the right eigenvector denominator should be b(1+ε), not b(1-ε), as confirmed by Eq. (15) and the phase boundaries. That is easy to fix. The more substantive caveat is the zero-frequency limit. The zero mode has no velocity damping at ω=0, so a monochromatic zero-frequency drive does not produce a stationary response; the paper regularizes with a finite frequency width η in Eq. (12). The stress-test note is right that the paper should be explicit that the observable is measured at a fixed nonzero drive frequency, where the response is K0/[ω²(ω²+γ²)], and the exponential growth of K0 with N is what makes the phase transition visible. This is not a fatal flaw—the prediction is well defined away from ω=0—but the paper would be strengthened by stating the protocol.\n\nThe diagonalizability assumption at exceptional points is another minor point. The bulk modes become defective at |ε|=1, but the zero mode appears to remain isolated; the paper should say a word about the limit, but I do not think it changes the result.\n\nCitation pattern looks fine. The paper builds on the experiment and the standard non-Hermitian skin-effect literature, and the Petermann-factor lineage is appropriate.\n\nBottom line: this is a useful theoretical contribution for people working on non-Hermitian mechanics and sensing. It deserves peer review and would benefit from a minor revision that fixes the typo, adds the measurement-frequency protocol, and flags the exceptional-point limit.\n\nBest,\n[Your name]","headline":"A clean, testable prediction that the zero-mode skin effect in nonreciprocal metamaterials comes with an exponentially divergent Petermann factor; minor fixes needed but deserves peer review.","tokens_in":8452,"tokens_out":5931,"would_cite":true,"duration_ms":53886,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper predicts a response phase transition in nonreciprocal mechanical metamaterials: inside the zero-mode skin-effect window the low-frequency response diverges with system size.","keywords":["nonhermitian skin effect","nonreciprocal metamaterials","zero modes","Petermann factor","biorthogonality","topological mechanics","exceptional points","response theory"],"falsifier":"Take the $N$-oscillator chain with $a=1$, $b=0.73$, $\\varepsilon=0.3$ (inside the window $0.156<\\varepsilon<6.41$) and drive it at zero frequency with a narrow frequency width $\\eta$; if the total power $P^{\\rm tot}(0)$, or the $K_0$ extracted from the line shape via Eq. (14), does not grow roughly exponentially as $N$ is increased (say $N=5,9,13,17$), the predicted response phase transition is falsified.","tokens_in":7489,"feed_emoji":"⚙️","tokens_out":13746,"duration_ms":109588,"temperature":0.7,"pith_summary":"This paper predicts that the nonhermitian skin effect of a topological zero mode in a nonreciprocal mechanical metamaterial is accompanied by a response phase transition: inside the coupling window $\\varepsilon_1<\\varepsilon<\\varepsilon_2$, the system becomes exponentially more sensitive to low-frequency excitation as it grows. The quantitative object is the Petermann factor $K_0$, a condition number measuring the nonorthogonality of the right and left zero-mode eigenvectors; it remains finite outside the window and diverges exponentially with system size inside it. The transition is not an exceptional-point effect: the resonance spectrum stays real across the phase, and the sensitivity does not peak at the bulk exceptional point. Because the response's input profile is set by the left eigenvector and its output profile by the right eigenvector, the power spectrum makes the underlying biorthogonality directly observable, and the total response exposes the skin-effect phase as a high-sensitivity regime requiring no fine-tuning.","feed_headline":"Zero-mode skin effect opens a phase of diverging response","feed_subtitle":"Chains become critically sensitive to low-frequency drive across the whole skin-effect window, no fine-tuning.","key_machinery":"The central object is the Petermann factor $K_{\\bar k}=(U^\\dagger U)_{\\bar k\\bar k}(U^{-1}U^{-\\dagger})_{\\bar k\\bar k}$, a condition number that quantifies how nonorthogonal the right and left eigenvectors of mode $\\bar k$ are; for the zero mode it reduces to Eq. (15), a ratio of geometric sums. The argument runs through the biorthogonal spectral decomposition of the Green's function $\\hat G=(\\omega^2\\mathbb{1}-M)^{-1}$: the left eigenvectors determine how strongly a drive at each site couples to the mode, the right eigenvectors determine where the mode's response appears, and $K_{\\bar k}$ is the product of their squared norms that weights the total response near resonance. The same machinery yields the separate input and output power spectra $P^{\\rm in}_m$ and $P^{\\rm out}_n$, which map out the left and right eigenvectors individually.","core_discovery":"For the chain whose dynamical matrix factors as $M=QR$, the right zero mode is $u_{0,n}\\propto [a(1-\\varepsilon)/(b(1+\\varepsilon))]^n$ and the left zero mode is $v_{0,n}\\propto (a/b)^n$; the right mode switches edge at $\\varepsilon_1=(a-b)/(a+b)$ and $\\varepsilon_2=(a+b)/(a-b)$, while the left mode stays put. The paper shows that the total power spectrum near zero frequency is $P^{\\rm tot}(\\omega)\\approx K_0/(\\omega^4+\\omega^2\\gamma^2)$, with $K_0$ given by Eq. (15), a ratio of geometric sums in the two localization lengths. In the thermodynamic limit $K_0$ stays finite outside the skin-effect phase and grows exponentially in $N$ for $\\varepsilon_1<\\varepsilon<\\varepsilon_2$. The paper's central discovery is that this response phase transition is a generic consequence of the zero-mode skin effect: it occurs over the entire phase, is independent of spectral singularities, and reveals the biorthogonal structure in a directly measurable response.","pith_inferences":["If the divergence is observable, the skin-effect window becomes a ready-made sensing phase: operate anywhere inside it, drive at the edge selected by the left eigenvector, and read the amplified response at the edge selected by the right eigenvector, with no exceptional-point fine-tuning.","The same $K_0$ that weights the driven response should also weight spontaneous fluctuations, so the phase transition might be detectable passively through excess noise rather than by external driving.","A finite-$N$ experiment should show $K_0$ rising steeply as $\\varepsilon$ crosses $\\varepsilon_1$ and falling after $\\varepsilon_2$, with the maximum inside the phase but away from the bulk exceptional point; any other pattern would indicate an extra ingredient beyond the biorthogonal response theory.","Since $K_0$ depends only on the two localization lengths, any nonreciprocal chain whose zero-mode profile is geometric in $a/b$ and $a(1-\\varepsilon)/b(1+\\varepsilon)$ should exhibit the same phase transition, regardless of the microscopic feedback mechanism."],"forward_implications":["For any $\\varepsilon$ in $\\varepsilon_1<\\varepsilon<\\varepsilon_2$, the zero-mode Petermann factor $K_0$ grows exponentially with system size, so the low-frequency response of a sufficiently large nonreciprocal chain diverges.","The enhanced sensitivity is a phase property: it holds across the whole skin-effect window, with no need to tune to an exceptional point, and the resonance spectrum remains real.","A position-resolved power-spectrum measurement separates the two biorthogonal partners: the input spectrum follows the left eigenvector, the output spectrum follows the right eigenvector.","The onset and end of the sensitive phase coincide with the relocalization of the right zero mode at $\\varepsilon_1$ and $\\varepsilon_2$, so response measurements pin down the zero-mode skin-effect boundaries.","Because the response formulas hold for arbitrary linear dynamical matrices, the same transition should appear in other nonreciprocal nonhermitian media, including skin effects of nonzero modes and topoelectric circuits."],"supporting_citations":[{"why":"supplies the nonreciprocal robotic-metamaterial platform with directed couplings that the response theory is built around.","marker":"[1]"},{"why":"provides the experimental zero-mode skin effect and the parameter values ($a=1$, $b=0.73$) used for the quantitative predictions.","marker":"[2]"},{"why":"introduces the nonhermitian localization by a flux imbalance that underlies the skin effect.","marker":"[3]"},{"why":"establishes the biorthogonal bulk-boundary correspondence that motivates tracking right and left eigenvectors together.","marker":"[6]"},{"why":"defines the Petermann factor as the excess-noise and enhanced-sensitivity measure in nonhermitian systems.","marker":"[12]"},{"why":"connects the Petermann factor to mode nonorthogonality and degeneracies, the statistical interpretation used here.","marker":"[14]"},{"why":"supplies the hermitian Kane-Lubensky topological mechanics model whose factorization $M=QQ^T$ is modified to $M=QR$.","marker":"[16]"},{"why":"provides the condition-number interpretation of $K\\ge 1$ as a measure of eigenvector nonorthogonality.","marker":"[20]"},{"why":"represents the exceptional-point sensing mechanism that the predicted phase transition does not require.","marker":"[24]"}],"fun_headline_variants":["Zero-mode skin effect yields diverging response without fine-tuning","Biorthogonality emerges in measurable response of mechanical chains","Skin effect of zero modes triggers a response phase transition","Diverging response exposes biorthogonality of zero modes"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing assumption is that a real experimental drive can be modeled by the harmonic equation (5) with velocity-dependent damping $\\gamma$ plus a finite frequency width $\\eta$ for the zero mode; if the physical excitation has no such width, the formally diverging Petermann factor may not translate into a diverging observable response.","fun_headline_variants_meta":{"raw":{"variants":["Zero-mode skin effect yields diverging response without fine-tuning","Biorthogonality emerges in measurable response of mechanical chains","Skin effect of zero modes triggers a response phase transition","Diverging response exposes biorthogonality of zero modes"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001143,"raw_usage":{"total_tokens":4738,"prompt_tokens":937,"completion_tokens":3801,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":553,"completion_tokens_details":{"reasoning_tokens":3732}},"tokens_in":553,"tokens_out":3801,"duration_ms":22652,"temperature":1.0,"reasoning_tokens":3732,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T12:50:17.957809+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take the $N$-oscillator chain with $a=1$, $b=0.73$, $\\varepsilon=0.3$ (inside the window $0.156<\\varepsilon<6.41$) and drive it at zero frequency with a narrow frequency width $\\eta$; if the total power $P^{\\rm tot}(0)$, or the $K_0$ extracted from the line shape via Eq. (14), does not grow roughly exponentially as $N$ is increased (say $N=5,9,13,17$), the predicted response phase transition is falsified.","supporting_citations":[{"cited_title":"Non-reciprocal robotic metamaterials","cited_arxiv_id":"1903.03807","evidence_quote":"supplies the nonreciprocal robotic-metamaterial platform with directed couplings that the response theory is built around."},{"cited_title":"Observation of non-Hermitian topology and its bulk-edge correspondence in an active mechanical metamaterial","cited_arxiv_id":"1907.11619","evidence_quote":"provides the experimental zero-mode skin effect and the parameter values ($a=1$, $b=0.73$) used for the quantitative predictions."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"defines the Petermann factor as the excess-noise and enhanced-sensitivity measure in nonhermitian systems."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"connects the Petermann factor to mode nonorthogonality and degeneracies, the statistical interpretation used here."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"supplies the hermitian Kane-Lubensky topological mechanics model whose factorization $M=QQ^T$ is modified to $M=QR$."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"provides the condition-number interpretation of $K\\ge 1$ as a measure of eigenvector nonorthogonality."}],"review_version":1}