{"id":"3466aa34-7f87-4df0-bdf5-c3c44ede63fb","arxiv_id":"1908.05439","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"A two-port drive with adjustable phase and amplitude continuously steers a cavity-magnon polariton between level repulsion and level attraction.","lead":"This paper shows that driving a cavity-magnon system through two microwave inputs with a controllable relative phase and amplitude allows continuous tuning between level repulsion and level attraction. It derives a new scattering formula and demonstrates intermediate and high amplitude regimes in a YIG-cavity experiment.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Eq. (5) rests on an asserted non-Hermitian a†m term; Eq. (4) is never derived from a two-port input-output model or quantitatively fitted.","rationale":"The paper has genuine strengths: it identifies a two-port control parameter, demonstrates qualitative crossover from repulsion to attraction in Fig. 5, explicitly discusses crosstalk limits, and the g'(δ,φ) expression has a clear algebraic origin once the non-Hermitian Hamiltonian is granted. I am not disputing that a non-Hermitian effective description can produce level attraction; this is standard in the dissipative-coupling literature. The problem is specifically that the effective Hamiltonian is not derived from the stated input-output formalism or from the physical torque mechanism, and the resulting Eq. (4) is never quantitatively tested. In good faith, one can read Sec. 3.2 as a phenomenological proposal; if so, the claim 'we derive a new expression from Input-Output theory' (abstract) overstates the support. The b_in2/b_in1=δ0e^{iφ} identification is a second, separate gap, but it affects mainly the third term of Eq. (4) rather than the pole condition; therefore I weight the non-Hermitian Hamiltonian assumption as the more load-bearing of the two. The reader's weakest assumption overlaps with mine only partially because the reader focuses on the amplitude-ratio mapping, while the deeper issue is how δ0 enters the coupling at all. Because the argument is plausible and the qualitative data support the existence of the effect, the appropriate verdict remains conditional rather than reject. An independent derivation or a quantitative fit would upgrade it; absence of either leaves the central formula under-supported.","tokens_in":15484,"tokens_out":11530,"duration_ms":110684,"concrete_test":"Independently re-derive Eq. (4) from the two-port Langevin equations (Eq. (2) plus a direct magnon drive sqrt(2κ_e2)b_in2) using only the Hermitian Tavis-Cummings coupling, and also using a dissipative-coupling model (imaginary off-diagonal Γ, as in Ref. [25]) instead of the a†m-only non-Hermitian term. If neither route yields the pole condition -i(ω-ω_c)+κ_c + g_eff^2(1+δ0 e^{iφ})/(-i(ω-ω_m)+κ_m)=0, then Eq. (5) is not a derived consequence of two-port driving. Additionally, fit Eq. (4) to the complex S11 spectra of Fig. 5 with κ_c, κ_m, g_eff fixed by the single-port limit; if the extracted g' deviates from Eq. (5) beyond uncertainties, the central formula is falsified.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central formula g'(δ0,φ)=g_eff sqrt(1+δ0 e^{iφ}) is read off the pole condition of Eq. (4), where the denominator contains g_eff^2(1+δ0 e^{iφ})/X. This replacement of the standard Hermitian coupling by a complex factor is not a consequence of the input-output equations in Sec. 3.1; it is inserted by positing, in Sec. 3.2, the non-Hermitian Hamiltonian with the additional term ħg_effδ0e^{iφ}a†m and deliberately omitting its Hermitian conjugate. The stated justification—that the conjugate 'would correspond to the crosstalk'—does not follow: crosstalk is direct port-to-port leakage, whereas the omitted m†a terms would modify the coherent cavity-to-magnon coupling. The paper provides no independent derivation of this non-Hermitian term from the torque mechanism of Sec. 4.3 or from standard coupled-mode theory. Moreover, the third term in Eq. (4) requires setting b_in2/b_in1=δ0e^{iφ}; δ0 is experimentally calibrated as an internal AC-field ratio through a circle-fit factor ζ, and the equality of these two ratios is asserted rather than derived. Because Eq. (5) is the entire basis for the claimed continuous control from repulsion to attraction, and because the experimental spectra (Figs. 5 and 6) are presented qualitatively without fits to Eq. (4), the central claim is not yet independently secured.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript studies a two-port driven cavity-magnon-polariton system in which a second microwave port couples directly to the magnons. It extends the single-port input-output treatment by proposing a non-Hermitian Hamiltonian that adds the term ħ g_eff δ0 e^{iφ} a†m and omits its Hermitian conjugate. From this model the authors derive a reflection coefficient S11(ω) (Eq. (4)) and introduce an effective complex coupling g′(δ0,φ) = g_eff sqrt(1 + δ0 e^{iφ}) (Eq. (5)). They predict level merging at δ0 = 1, φ = π and level attraction for δ0 > 1, φ = π, and they present numerical plots of the real and imaginary parts of g′ as functions of δ0 and φ. Experimentally, they report spectra showing intermediate phases between attraction and repulsion (Fig. 5) and a high-δ0 measurement with δ0 = 11.79 ± 1.97 (Fig. 6), discussing crosstalk limitations. The framing is that the relative phase and amplitude of the second port provide broad in-situ control over the coherent information exchange between cavity photons and magnons.","tokens_in":15929,"tokens_out":6747,"duration_ms":68165,"significance":"If the central formula and mechanism are correct, the work would be a valuable step toward in-situ control of the coherent versus dissipative character of cavity-magnon coupling, including intermediate coexistence regimes. The paper also deserves credit for clearly identifying crosstalk as a practical limitation, for emphasizing the role of the tilted magnon-port geometry, and for extending the authors' earlier level-merging observation into intermediate-phase and high-δ0 regimes. However, the central theoretical step is an asserted non-Hermitian Hamiltonian rather than a derived input-output or coupled-mode result, and the experimental spectra are not quantitatively fitted to Eq. (4). The significance of the claim is therefore conditional on a derivation and a quantitative comparison that are not yet present.","major_comments":[{"comment":"The central formula hinges on adding the term ħ g_eff δ0 e^{iφ} a†m and omitting its Hermitian conjugate, with the statement that the conjugate 'would correspond to the crosstalk.' This is not a derivation: crosstalk is direct port-to-port microwave leakage, whereas the omitted m†a term would change the coherent cavity–magnon exchange. The equations of motion consequently contain asymmetric off-diagonal terms (−i g_eff a in dm/dt, −i g_eff(1+δ0 e^{iφ}) m in da/dt) that are posited rather than obtained from a microscopic torque or coupled-mode calculation. Because Eq. (5) is read directly from the denominator generated by this asymmetric Hamiltonian, the main prediction is not independently established. Please derive the effective coupling from a standard two-port input-output treatment or from the Landau-Lifshitz torque mechanism, and specify how the bath and crosstalk terms are treated.","section":"§3.2, Hamiltonian before Eq. (4)"},{"comment":"The third term of Eq. (4) implicitly uses b_in2/b_in1 = δ0 e^{iφ}, but δ0 is defined in §4.2 as the ratio of internal AC magnetic fields at the sample, obtained from external amplitudes through a calibration factor ζ. The equality of the external-input ratio and the internal-field ratio is not derived. Different ports have different mode overlaps and coupling efficiencies, and a complex calibration factor could enter. Moreover, Eq. (4) as written contains no explicit b_in2/b_in1 ratio, so it is unclear how the third term was normalized. Without this mapping, Eq. (4) is not a closed expression and the quantitative prediction of level merging at δ0 = 1, φ = π is not justified.","section":"§3.2, Eq. (4)"},{"comment":"The effective coupling g′ = g_eff sqrt(1 + δ0 e^{iφ}) is obtained by replacing g_eff² with g_eff²(1 + δ0 e^{iφ}) in the reflection denominator. This is a restatement of the model rather than an independent consequence of input-output theory. It is also incomplete: the full expression Eq. (4) contains a third term proportional to 2i g_eff δ0 e^{iφ}(1 + δ0 e^{iφ}) sqrt(κ_e1 κ_e2) divided by X times the denominator. The pole structure of Eq. (4) is not computed. A claim that complete merging occurs exactly at δ0 = 1, φ = π should be checked against the full denominator of Eq. (4), including κ_e2 and crosstalk contributions, not only against the simplified factor in Eq. (5).","section":"§4.1, Eq. (5)"},{"comment":"The experimental validation is qualitative. No fits to Eq. (4) are reported, no extracted values of Re g′ or Im g′ are shown as functions of φ or δ0, and the quoted uncertainties in δ0 are not propagated into the claimed coupling behavior. The asserted coexistence of repulsion and attraction in Fig. 5 is based on visual inspection of line shapes and phase jumps, and in Fig. 6 the level-merging signal and the crosstalk anticrossing are separated by eye. A quantitative fit of the full S11 expression to all spectra would directly test Eq. (5) and is needed to support the central claim of broad, quantitative tunability.","section":"§4.4 and §4.5, Figs. 5 and 6"}],"minor_comments":[{"comment":"The axis labels and panel annotations contain corrupted characters (e.g., '/uni00000003/...'), making the plots difficult to read; please regenerate them with clean LaTeX labels.","section":"Fig. 3"},{"comment":"The displayed formula has unbalanced parentheses in the denominator of the third term and appears malformed; please correct the typography.","section":"Eq. (4)"},{"comment":"There are operator-ordering and prefactor typos: the cavity term should be a†a rather than aa†, and the magnon number term is missing the factor ħ.","section":"§3.1, Eq. (2) and preceding Hamiltonian"},{"comment":"The main text contains a stray 'ß If hAC...' passage, and several instances of 'e.f.' should read 'e.g.'; please proofread the text.","section":"§4.3"},{"comment":"The statement that crosstalk 'has to be considered in the calculation of ℑ(g′(δ0,φ))' is not accompanied by an explicit formula; please give the concrete procedure used to extract the imaginary part of the coupling from the measured spectra.","section":"§4.5"}],"recommendation":"major_revision","confidential_remarks":"The main concern is not novelty but the missing derivation and the absence of quantitative fits. I would encourage the editor to request a derivation of the non-Hermitian coupling from a standard coupled-mode or torque model, and fits of Eq. (4) to the reported spectra. The PT-symmetry and exceptional-point discussion in Sec. 5 is loosely connected to the rest of the paper and should be made precise or removed."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a real experimental step beyond the authors' own Ref. 28, but the central theoretical result is not secured. Eq. (5) is presented as a new expression, yet it is simply the pole condition of a Hamiltonian in which the complex term hbar g_eff delta0 e^{i phi} a-dagger m was inserted by hand. The stress-test note holds up on reading: the justification that the omitted m-dagger a term 'would correspond to the crosstalk' conflates direct port-to-port leakage with the coherent cavity-magnon coupling. And the mapping from external drive amplitudes to the internal field ratio delta0 is asserted, not derived. So Eq. (5) is a restatement of the model, not an independent prediction.\n\nWhat is genuinely useful: the paper writes down an S11 expression for the two-port configuration, goes beyond Ref. 28 by showing intermediate phases and high-delta0 data, and is honest about crosstalk and signal-to-noise limits. The gradual transition between repulsion and attraction in Fig. 5 looks like a real effect. The high-delta0 spectra are presented as guides, not fits.\n\nWhere it falls short: no quantitative fitting of the spectra to Eq. (4), no extraction of g' from data, and no microphysical derivation of the non-Hermitian term from the torque/dissipative-coupling mechanism suggested in Sec. 4.3. The paper connects to known dissipative-coupling models by analogy, not by derivation. So the claim of 'broad tunability' is plausible but only qualitatively validated.\n\nThe citation pattern is fine; Ref. 28 is properly built on. This is an incremental experimental contribution to cavity spintronics, likely useful to people working on level attraction and PT-symmetric magnonics. It deserves referee time, but a serious referee should insist on a cleaner derivation and quantitative analysis before publication.","headline":"A plausible two-port cavity-magnon control knob, but Eq. (5) is read off an ad hoc non-Hermitian Hamiltonian and the experiments are only qualitative; worth refereeing, not desk-rejecting.","tokens_in":16346,"tokens_out":3085,"would_cite":true,"duration_ms":32539,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Adding a second microwave port turns the cavity-magnon coupling strength complex, so a single phase knob sweeps the system from level repulsion through complete level merging into level attraction.","keywords":["cavity magnon-polaritons","level repulsion","level attraction","non-Hermitian Hamiltonian","two-port microwave driving","coupling strength control","magnon-photon coupling","input-output theory"],"falsifier":"Measure the dispersion of the two-port system with independently calibrated internal AC fields at the sample position, for instance by locally probing the microwave magnetic field, and check whether the gap closes at exactly $\\delta_0=1$ and $\\phi=\\pi$; any systematic shift or residual splitting at that point would falsify the formula $g'=g_{\\mathrm{eff}}\\sqrt{1+\\delta_0 e^{i\\phi}}$.","tokens_in":15304,"feed_emoji":"🧲","tokens_out":6148,"duration_ms":54002,"temperature":0.7,"pith_summary":"This paper claims that adding a second microwave input to a cavity-magnon system changes the effective light-magnon coupling from a real number into a complex one, $g'(\\delta_0,\\phi)=g_{\\mathrm{eff}}\\sqrt{1+\\delta_0 e^{i\\phi}}$, set by the relative amplitude $\\delta_0$ and phase $\\phi$ of the two drives. Because the real part of the coupling produces level repulsion while the imaginary part produces level attraction, tuning $\\phi$ and $\\delta_0$ sweeps the system continuously between the two regimes. At $\\delta_0=1$ and $\\phi=\\pi$ the gap closes completely, which the paper calls level merging. The authors derive a new reflection formula from input-output theory, verify the predicted coexistence of repulsion and attraction at intermediate phases, and push the amplitude ratio to $\\delta_0\\approx 11.8$, where crosstalk becomes the limiting factor. If correct, this gives in-situ, continuous control of how strongly cavity photons and magnons exchange information.","feed_headline":"Two-port drive tunes magnon coupling from repulsion to attraction","feed_subtitle":"Relative phase and amplitude between the two ports turn the coupling complex, closing the anticrossing gap at a single working point.","key_machinery":"The load-bearing object is the complex effective coupling $g'(\\delta_0,\\phi)=g_{\\mathrm{eff}}\\sqrt{1+\\delta_0 e^{i\\phi}}$, obtained from a non-Hermitian Hamiltonian in which the magnon-port drive adds a term $\\hbar g_{\\mathrm{eff}}\\delta_0 e^{i\\phi} a^\\dagger m$ without its Hermitian conjugate, the latter being identified with unwanted crosstalk. This single formula carries the argument: its real part is assigned to level repulsion, its imaginary part to level attraction, and its zero at $\\delta_0=1,\\phi=\\pi$ produces level merging. The same term enters the derived reflection coefficient through the substitution $g_{\\mathrm{eff}}^2 \\to g_{\\mathrm{eff}}^2(1+\\delta_0 e^{i\\phi})$ plus a second contribution proportional to the magnon-port coupling. Physically the mechanism is an additional torque on the magnetization that, depending on phase and amplitude, compensates or overdrives the dissipative channels, moving the system between coherent and dissipative coupling regimes.","core_discovery":"The central claim is that the two-port drive makes the coupling strength complex: $g'(\\delta_0,\\phi)=g_{\\mathrm{eff}}\\sqrt{1+\\delta_0 e^{i\\phi}}$, where $\\delta_0$ is the ratio of the AC magnetic fields at the magnon port and cavity port and $\\phi$ their relative phase. For $\\phi=0$ the coupling stays real and grows with $\\delta_0$, so the spectrum keeps its avoided crossing. For $\\phi=\\pi$ the real part vanishes once $\\delta_0\\ge 1$; the coupling becomes purely imaginary, which is the signature of level attraction, and at $\\delta_0=1,\\phi=\\pi$ the gap closes entirely (level merging). At intermediate phases both real and imaginary parts are present, so repulsion and attraction coexist in one spectrum. Experimentally the paper observes that coexistence and, at high $\\delta_0$ with $\\phi=\\pi$, a broadened coalesced region whose width grows with $\\delta_0$, limited at $\\delta_0\\approx 11.8$ by crosstalk. The paper also identifies the microscopic mechanism as a transition from coherent coupling to dissipative coupling: the tilted magnon port produces an AC field component along the effective field that modulates the magnon frequency, detuning it from the cavity photon.","pith_inferences":["Beyond the paper, the identity $g'=g_{\\mathrm{eff}}\\sqrt{1+\\delta_0 e^{i\\phi}}$ suggests that the two-port drive engineers a synthetic imaginary coupling; a direct test would be to extract the complex phase of the reflection coefficient as a function of $\\phi$ and compare it with the predicted argument of $g'$.","Beyond the paper, the non-Hermitian structure hints at an exceptional point: fixing $\\delta_0=1$ and sweeping $\\phi$ through $\\pi$ should make the real frequency splitting vanish while the eigenmodes coalesce, which could be probed experimentally with the same two-port setup.","Beyond the paper, crosstalk at high $\\delta_0$ acts as a parasitic real coupling; a natural extension is to design a compensating orthogonal coupler for the magnon port to suppress this term and push further into the level-attraction regime.","Beyond the paper, the same complex-coupling mechanism should apply to other hybrid systems with two coherent drives, provided the second port couples to only one subsystem, so the result could transfer to optomechanical or superconducting-circuit platforms."],"forward_implications":["At $\\phi=\\pi$ and $\\delta_0$ just above 1, the gap closure widens into a finite coalesced region; the paper observes about 0.5 mT of width at $\\delta_0=11.8$.","Intermediate phases allow continuous control of the relative weight of repulsion and attraction in the same spectrum, which the paper describes as a way to set the transmitted information flow between cavity photon and magnon.","The complex coupling formula implies that the scattering parameter contains both real and imaginary contributions, so phase-resolved measurements are needed to identify level attraction reliably, especially at high $\\delta_0$.","The two-port control requires no mechanical changes to the resonator, so in-situ tuning could be transferred to cryogenic or quantum-coherent settings, such as coupling to a superconducting circuit."],"supporting_citations":[{"why":"Provides the input-output formalism and Langevin equations used to derive the single-port reflection formula that the two-port result extends.","marker":"[21]"},{"why":"Supplies the theory of the additional torque on the magnetization from a second drive, which is the physical mechanism behind the coupling control.","marker":"[25]"},{"why":"Reports the authors' earlier two-port experiment that first observed level merging and defines the setup being generalized here.","marker":"[28]"},{"why":"Gives the microscopic model of coherent versus dissipative coupling used to explain level attraction through magnon-frequency detuning.","marker":"[38]"},{"why":"Demonstrates level attraction by sample repositioning in a single-port system, serving as the alternative approach the paper contrasts with two-port driving.","marker":"[24]"},{"why":"Connects non-Hermitian cavity magnon-polaritons to PT symmetry and exceptional points, framing the spectral consequences discussed in the outlook.","marker":"[20]"},{"why":"Supplies the circle-fit calibration method used to extract the internal amplitude ratio $\\delta_0$ from external drive amplitudes.","marker":"[37]"},{"why":"Provides the standard transmission measurement and Rabi-splitting description of single-port level repulsion that forms the baseline for the two-port comparison.","marker":"[4]"}],"fun_headline_variants":["Two-port phase twists magnon coupling into attraction","Complex coupling steers magnon polaritons to level attraction","Two-port drive closes the gap from repulsion to attraction","Phase and amplitude rewire magnon coupling between repulsion and attraction"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The derivation assumes that the second port's effect is fully captured by setting the ratio of the drive amplitudes entering the equations to $\\delta_0 e^{i\\phi}$, with $\\delta_0$ equal to the internal AC-field ratio, while the omitted Hermitian-conjugate term is exactly the crosstalk and can be dropped; if the mapping from external amplitudes to internal fields is wrong, the central formula for the coupling strength does not follow.","fun_headline_variants_meta":{"raw":{"variants":["Two-port phase twists magnon coupling into attraction","Complex coupling steers magnon polaritons to level attraction","Two-port drive closes the gap from repulsion to attraction","Phase and amplitude rewire magnon coupling between repulsion and attraction"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000901,"raw_usage":{"total_tokens":3928,"prompt_tokens":1046,"completion_tokens":2882,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":662,"completion_tokens_details":{"reasoning_tokens":2814}},"tokens_in":662,"tokens_out":2882,"duration_ms":20447,"temperature":1.0,"reasoning_tokens":2814,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:14:16.093529+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the dispersion of the two-port system with independently calibrated internal AC fields at the sample position, for instance by locally probing the microwave magnetic field, and check whether the gap closes at exactly $\\delta_0=1$ and $\\phi=\\pi$; any systematic shift or residual splitting at that point would falsify the formula $g'=g_{\\mathrm{eff}}\\sqrt{1+\\delta_0 e^{i\\phi}}$.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the input-output formalism and Langevin equations used to derive the single-port reflection formula that the two-port result extends."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Demonstrates level attraction by sample repositioning in a single-port system, serving as the alternative approach the paper contrasts with two-port driving."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the circle-fit calibration method used to extract the internal amplitude ratio $\\delta_0$ from external drive amplitudes."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the standard transmission measurement and Rabi-splitting description of single-port level repulsion that forms the baseline for the two-port comparison."}],"review_version":1}