{"id":"e93a2aa8-f5f0-40bd-89ec-5d7599274db9","arxiv_id":"2412.04544","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"A Dirac spin liquid with random-phase-approximation corrections reproduces the phase diagram and spin dynamics, including THz and neutron signatures, of the frustrated honeycomb J1-J3 XY model relevant to cobaltates.","lead":"This paper argues that the weak magnetic order seen in honeycomb cobaltate materials grows out of a nearby quantum spin liquid, a state with no magnetic order whose excitations are fractional. Using numerical and analytical methods, the authors compute the spin dynamics of this 'parent' state and match key features of recent THz and neutron scattering experiments.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"RPA/gauge-fluctuation suppression is the weak link: Eq. (A20) has no small parameter at the working point, and Fig. 3 is tuned via alpha3/alpha1 and alpha1, so the response agreement is not yet a controlled prediction.","rationale":"The reader's weakest assumption is the same one I identify: the uncontrolled parton/RPA decomposition, together with the tuning of alpha to match previous numerics, is the load-bearing fragility of the paper. The central claim has two independent legs: (i) the VMC energy competition showing a proximate DSL and weak orders within 1-2%, and (ii) the RPA instability and response calculation compared to experiments. The first leg is genuine evidence and I do not object to it. The second leg is what supports the dynamical claim, and it is exactly the part the paper itself concedes is uncontrolled: Sec. I says the authors do not aim for a controlled calculation, Sec. VII lists a controlled diagrammatic route as future work, and the monopole-confinement remark in Sec. I applies to the ordered phases where the experimental data are taken. Eq. (A20) is obtained after discarding gauge fluctuations and truncating the Weiss-field action at quadratic order; at the working point the RPA coupling is not small, so there is no a priori guarantee that vertex corrections or gauge-boson exchange leave Figs. 6 and 12 qualitatively intact. The tuning of alpha in Sec. IV further weakens the predictive content of the phase diagram: matching DMRG by construction cannot count as a successful prediction. A concrete test is therefore to include the leading gauge/vertex correction; if the qualitative features survive, the concern is answered, and if not, the experimental agreement is parameter-sensitive. Because the paper is explicit about its limitations and the VMC leg is real, I would not change the reader's CONDITIONAL verdict.","tokens_in":20948,"tokens_out":7866,"duration_ms":84322,"concrete_test":"At the star point (J3/J1 = 0.35, alpha1 = 0.7, alpha3/alpha1 = 0.6), recompute chi_RPA(Q, omega) after including the leading gauge-field correction: couple the spinons to the phase fluctuation of the condensed w fields in Eq. (A9), add the resulting self-energy to the spinon Green's function in Eq. (10), and re-evaluate chi_RPA in Eq. (11). Compare chi''_xx(Q, omega) along Gamma -> M with Fig. 6(a). If the sharp mode and continuum survive at similar energies and weights, the suppression of gauge fluctuations is benign; if the spectrum is substantially renormalized, the RPA response is not a reliable test of the proximate-DSL scenario.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central dynamical claim rests on the effective action in Eqs. (5)-(7) and (A20)-(A22): the hopping-channel Hubbard-Stratonovich fields are assumed condensed so the U(1) gauge field is dropped, and the Weiss-field action is truncated at quadratic order. Neither step has a small parameter at the point used for Figs. 5, 6, 11, and 12 (alpha1 ~ 0.7, J3/J1 ~ 0.35): the residual interaction (1 - alpha1)J1 ~ 0.3J1 is several times the spinon hopping scale t1 ~ 0.13 alpha1 J1 ~ 0.09J1, so vertex corrections and gauge-boson exchange are not controlled. The paper acknowledges this in Sec. I and Sec. VII, and it also notes that monopole proliferation will confine the partons in ordered phases where the experimental data are taken. In addition, Sec. IV fixes alpha3/alpha1 = 0.6 and tunes alpha1 to reproduce previous DMRG boundaries, so the successful phase diagram is partly enforced rather than predicted. The VMC energy competition in Fig. 1 is genuine independent support for a proximate-DSL picture, but it does not validate the RPA dynamical response. If the sharp Gamma-to-M mode and the continuum in Fig. 6 are sensitive to gauge/vertex corrections or to the choice of alpha, the main experimental conclusion is not yet secured.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This manuscript proposes a phenomenological parton theory for the spin-1/2 J1-J3 easy-plane XY model on the honeycomb lattice, aiming to describe the weakly ordered honeycomb cobaltates such as BaCo2(AsO4)2. The Hamiltonian is split as H = (1-alpha)H + alpha H, with the first piece decoupled via Weiss/magnetization Hubbard-Stratonovich fields and the second via condensed spinon-hopping fields; after integrating out the spinons, the Weiss-field effective action is truncated at quadratic order (RPA), giving the susceptibility chi = (chi0^{-1} - (1-alpha)J)^{-1} (Sec. II and App. A). The paper contains two kinds of results. First, variational Monte Carlo with Gutzwiller-projected parton states and Jastrow factors shows that in the intermediate regime 0.32 < J3/J1 < 0.37 the pure easy-plane Dirac spin liquid and DSL+weak-order ansatzes (FM, zig-zag, double zig-zag, spiral) lie within 1-2% in energy, with in-plane ordered moments of about 0.2 Bohr magneton. Second, with alpha3/alpha1 fixed at 0.6 and alpha1 tuned so that the instability lines roughly match earlier DMRG/VMC phase diagrams, the RPA instabilities of the DSL give FM, zig-zag, and incommensurate spiral orders, and the RPA-corrected dynamical spin response qualitatively reproduces THz and neutron-scattering features, including the spin continuum, sharp Gamma-to-M modes, and easy in-plane polarization.","tokens_in":21215,"tokens_out":14552,"duration_ms":137936,"significance":"The value of the paper, if its claims hold, is twofold. The VMC study is an independent numerical assessment supporting the idea that the J1-J3 XY model sits near a proximate Dirac spin liquid in the intermediate coupling window; this is a genuine calculation with optimized variational parameters, and the narrow (1-2%) energy spread among competing states provides a concrete, checkable basis for the parent-state scenario. The RPA-response framework also offers a useful qualitative template for interpreting continuum-plus-sharp-mode spectra in weakly ordered magnets at a time when the experimental phenomenology of BaCo2(AsO4)2 is actively debated. The paper is commendably explicit about its limitations: it states in Sec. I that it does not aim for a controlled calculation, in Sec. IV that the RPA phase diagram is matched to previous numerics by construction, and in Secs. I and VII that monopole proliferation and gauge fluctuations are not treated. The honesty of these statements is a strength, but it also means that the headline dynamical results (Figs.","major_comments":[{"comment":"The headline RPA phase diagram is calibrated rather than predicted. The text fixes alpha3/alpha1 = 0.6 and tunes alpha1 'such that the obtained RPA phase diagram roughly matches the results of previous numerics'; because the same {alpha} set the effective interaction kernel J(Q;{alpha}) in Eq. (A20) and hence both the instability lines and the RPA response functions, the agreement of Fig. 3 with DMRG [55, 59] is in part enforced by construction. This is a legitimate phenomenological strategy, and the manuscript states it honestly in Sec. IV, but the abstract's claim of a 'phase diagram which is consistent with VMC and DMRG studies' and Sec. VII's 'reproduce the full many-body phase diagram' overstate what was done. The fix is to (i) explicitly label which results are calibrated and which are predictions, and (ii) provide a robustness check of the response functions (Figs. 5-6, 11-12) over the full range alpha3/alpha1 = 0.5-0.8 and alpha1 = 0.6-0.7 for which the instability lines agree with previous numerics, rather than at the single star point.","section":"Sec. IV, Fig. 3"},{"comment":"The RPA truncation and gauge-field suppression have no small parameter at the working point. At the response-calculation point (Sec. V: alpha1 ~ 0.7, J3/J1 ~ 0.35), the residual interaction (1-alpha1)J1 ~ 0.3J1 is several times the spinon hopping scale t1 ~ 0.13 alpha1 J1 ~ 0.09J1, so truncating the Weiss-field action at quadratic order and treating the magnetization field as weak in App. A are not justified by any obvious expansion parameter; vertex corrections and spinon self-energies are expected to be comparable to the terms kept. In addition, Eq. (5) drops the spatial gauge field by assuming the w fields are condensed, a significant step for a compact U(1) DSL, and the paper itself notes the danger of monopole proliferation (Sec. I). Since exactly these truncations are used for the central experimental comparisons (THz and INS, Figs. 5-6 and 11-12), the manuscript should provide a quantitative diagnostic of the truncation error (for example, a one-loop self-energy or vertex-correction estimate at the star point) or explicitly restrict the central claims to a qualitative demonstration.","section":"App. A, Eqs. (A20)-(A22); Secs. II and V"},{"comment":"The partonic interpretation of the in-field response is undermined by the confinement argument made in the same paper. The applied in-plane field gaps the Dirac nodes (Sec. VI C and App. C, Fig. 16b), and Sec. I states that in symmetry-broken phases that gap out the partons 'the monopoles of the gauge field will proliferate and lead to confinement'. The field-induced ordered regime of Figs. 10-12 is precisely such a gapped phase, so by the authors' own reasoning the sharp Gamma-to-M mode in Fig. 12 is better viewed as a magnon of the magnetically ordered state than as a two-spinon bound state. The paper should state explicitly what its calculation adds in this regime relative to linear spin-wave theory, or should limit the parton-RPA interpretation to the zero-field proximate-DSL region.","section":"Secs. I and VI"},{"comment":"The parent-state claim rests on 1-2% energy differences among the pure DSL and the DSL+order ansatzes, and on the claim that several order patterns compete in the intermediate window, but the VMC energy curves are shown without statistical error bars or optimization-convergence information. Appendix B reports 10,000 thermalization plus 10,000 measurement sweeps and 250 stochastic-reconfiguration steps, but no uncertainty estimate for the energies (or for the ordering moments in Fig. 2) is given. Given that the intermediate-window statement is load-bearing for the proximate-DSL scenario and that different DMRG studies disagree within this window [55, 59], the manuscript should report the statistical errors and, if available, a check that the variational optimization has converged at the J3/J1 values shown in Fig. 1.","section":"Fig. 1 and App. B"}],"minor_comments":[{"comment":"The sentence 'starting in the pure quantum limit at alpha1 = 0, the DSL is stable as alpha1 is decreased towards the classical limit' conflicts with the definitions in Sec. II, where alpha = 1 is the quantum limit and alpha = 0 the classical limit, and also with the parallel sentence in Sec. VI B, which correctly says 'at alpha1 = 1'. This appears to be a typographical error and should be corrected.","section":"Sec. IV, first paragraph"},{"comment":"The comparison with the THz and INS data of Refs. [41, 14] is qualitative ('slightly different energies', 'good agreement'), but the experimental energy values and constant-energy slices are never quoted; giving the experimental energy ranges (for example, the THz frequency window and the INS energy slices) would allow the reader to judge the size of the claimed discrepancies.","section":"Sec. V, Figs. 5-6"},{"comment":"The mean-field polarization field (Bx/J1 ~ 0.1-0.15) exceeds the VMC result (Bx/J1 ~ 0.02) by about an order of magnitude; since both are computed with similar alpha-based parameters, the brief comment that mean-field theory underestimates internal Weiss fields should be expanded into a quantitative discussion, otherwise the alpha framework appears internally inconsistent across the two methods.","section":"Sec. VI A, Figs. 8-9"},{"comment":"There are several presentation issues: 'limitα = 0' in Sec. II should read 'the limit α = 0'; the alpha_ij notation is introduced in Sec. II but only defined on bonds in App. A; and the sublattice indices on the Green's functions in Eq. (10) are not defined before first use.","section":"Secs. II and V"},{"comment":"The 1/T1 result is computed at the bare mean-field level, not with RPA corrections; this should be stated in the main text rather than only being implicit from the calculation description, so that the T^3 power law is not attributed to the full RPA framework.","section":"Sec. V C"},{"comment":"The statement that the color-bar scales are 'in arbitrary units but consistent between the different plots' is ambiguous; please clarify whether the same relative scale is used across all panels and whether the intensities of different panels can be compared quantitatively.","section":"Captions of Figs. 6 and 12"}],"recommendation":"major_revision","confidential_remarks":"The paper is within the scope of a strong condensed-matter journal and the VMC part is a genuine contribution. The main risk is overclaiming: the abstract and discussion present the RPA phase diagram and response functions as reproductions of numerics and experiment, when the alpha parameters were chosen to achieve exactly that match. My recommendation of major revision rests on asking the authors to convert this into an explicit calibrated-versus-predicted distinction and to add robustness checks (response functions over the allowed alpha range, VMC energy error bars, and an RPA truncation diagnostic). If the authors are not willing to add the RPA robustness check, the claims in the abstract and Sec. VII should be scaled back to match what is actually demonstrated."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nYou should know this paper is a serious, honest attempt to push a proximate-Dirac-spin-liquid picture for honeycomb cobaltates, but the dynamical response part rests on an admittedly uncontrolled RPA with fitted parameters. The VMC part stands on its own.\n\nWhat is actually new: the RPA-corrected spin response (THz, neutron, NMR), the in-plane field phase diagram and magnetization, and an expanded set of variational ansatzes including double zig-zag and spiral. The VMC energy competition in Fig. 1 is a genuine calculation and supports the idea of a DSL parent state in the intermediate J3/J1 window. The paper also does something rare: it states its own limitations openly. It says it does not aim for a controlled calculation, and it flags monopole proliferation and gauge fluctuations as open issues.\n\nThe soft spots are real but mostly acknowledged. The RPA phase diagram in Fig. 3 is tuned: alpha3/alpha1 is fixed at 0.6 and alpha1 is adjusted so the instability lines match previous DMRG. So the phase diagram is not an independent prediction. The response functions are computed at one representative point near the instability, without error estimates or sensitivity scans over alpha. The residual interaction (1-alpha1)J1 ~ 0.3J1 is larger than the spinon hopping scale, so vertex corrections are not small; the stress-test note is right that no small parameter protects the RPA at the working point. That said, the paper never oversells this. It presents the approach as phenomenological, in the spirit of spin-fermion models for cuprates. The citation pattern looks fair; the prior DSL parent-state work is cited and the present results are clearly incremental.\n\nI do not think the central narrative collapses. The VMC parent-state result and the qualitative response features are independent enough to be worth taking seriously. But the experimental agreement should be graded as suggestive, not predictive. No code or data are shipped, which is a minor drag for reproducibility.\n\nWho is this for? People working on cobaltates or on proximate spin-liquid dynamics in weakly ordered magnets. It deserves a serious referee: it is a well-structured, honest paper with new calculations and a clear experimental hook. I would send it to review, but I would ask the referee to focus on whether the response functions are robust to alpha and whether the fitted phase diagram undermines the dynamical claims.\n\nMy take: worth engaging, conditionally.","headline":"Honest, useful proximate-DSL study for honeycomb cobaltates: VMC parent-state evidence is solid, but the RPA dynamics are fitted rather than predicted, so the experimental agreement is suggestive, not conclusive.","tokens_in":21828,"tokens_out":1707,"would_cite":true,"duration_ms":48238,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["75.25.aj","75.40.Gb","75.70.Tj"],"model":"deepseek-v4-flash","headline":"This paper claims that an easy-plane Dirac spin liquid is the proximate parent of the weakly ordered honeycomb cobaltates, with spinon response reproducing their THz and neutron spectra.","keywords":["easy-plane Dirac spin liquid","honeycomb cobaltates","J1-J3 XY model","frustrated magnetism","parton RPA","spin dynamics","BaCo2(AsO4)2","THz spectroscopy"],"falsifier":"A concrete observation that would settle the claim: measure the THz absorption of BaCo2(AsO4)2 below 0.2 THz, the range the paper identifies as experimentally unexplored (Sec. V A). At $J_3/J_1 \\approx 0.35$ the RPA-corrected model predicts an enhanced low-energy peak near $\\Omega/J_1 \\approx 0.1$; a spectrum with no such peak, or with a purely magnon-like gapped onset, would rule out the proximate-DSL RPA description.","tokens_in":20645,"feed_emoji":"🧲","tokens_out":11591,"duration_ms":108459,"temperature":0.7,"pith_summary":"This paper argues that the spin dynamics of weakly ordered honeycomb cobaltates, specifically BaCo2(AsO4)2, are best understood by starting from an easy-plane Dirac spin liquid (DSL) rather than from magnons. In a spin-1/2 J1-J3 XY model on the honeycomb lattice, variational Monte Carlo shows that between J3/J1 ≈ 0.32 and 0.37 a pure DSL and DSL-plus-weak-order states are within 1-2% of each other in energy. A modified parton theory, in which the Hamiltonian is split into a classical Weiss-field piece and a quantum spin-liquid piece with the spinon gauge fluctuations frozen, produces RPA instabilities toward ferromagnetic, zig-zag, and incommensurate spiral order. The same RPA-corrected susceptibility reproduces the qualitative features of THz and neutron scattering on BaCo2(AsO4)2 at zero field and in an in-plane field, including a sharp-mode-plus-continuum spectrum and easy polarization. If right, this makes the magnet's low-energy response a window onto proximate-DSL physics rather than pure magnon physics.","feed_headline":"A Dirac spin liquid is the parent of weakly ordered cobaltates","feed_subtitle":"A spin-liquid wavefunction with RPA corrections reproduces THz and neutron data on a cobaltate magnet.","key_machinery":"The engine of the argument is a modified parton theory: rewrite the spin operators as fermionic spinons, split the Hamiltonian on every bond as $(1-\\alpha)$ times a classical Weiss-field channel plus $\\alpha$ times a spinon hopping channel, and assume the hopping fields are condensed around their mean-field values so gauge fluctuations are suppressed. Integrating out the spinons leaves a quadratic action for the Weiss fields whose RPA susceptibility is $\\chi = (1 - J\\chi_0)^{-1}\\chi_0$, where $J$ is the rescaled exchange and $\\chi_0$ is the bare spinon bubble. The easy-plane Dirac spin liquid is the mean-field state with opposite-signed spin-$\\uparrow$ and spin-$\\downarrow$ hoppings $t_1 \\approx 0.13J_1$, $t_3/t_1 \\approx 0.1$, which has Dirac nodes at $K$ and $K'$ and a nearly flat band edge whose van Hove peak near $0.1J_1$ generates the enhanced low-energy THz response. The RPA formula turns spinon fluctuations into both the sharp low-energy modes and the continuum seen in experiment.","core_discovery":"The central discovery claimed is that an easy-plane Dirac spin liquid is a viable parent state for the competing magnetic orders of the honeycomb J1-J3 XY model, and that this parent state, once perturbed by residual interactions at RPA level, explains the observed dynamics of honeycomb cobaltates. The evidence is variational: optimized Gutzwiller-projected wavefunctions place the pure DSL within 1-2% of the best DSL-plus-order states (ferromagnet, zig-zag, double zig-zag, spiral) in the window 0.32 < J3/J1 < 0.37, with ordered moments of order 25-50% of the full moment. The theoretical mechanism is a Hamiltonian split $H = (1-\\alpha)H + \\alpha H$, with the hopping channel condensed and the Weiss-field channel treated to quadratic order; the resulting RPA instabilities of the DSL give the same ferromagnet/zig-zag/incommensurate-spiral sequence as DMRG, and the RPA-corrected spin response shows sharp spinon-bound-state modes plus a spinon continuum that reproduce the shape of the THz and neutron data on BaCo2(AsO4)2. The paper does not claim a controlled calculation.","pith_inferences":["(Editorial inference) If the proximate-DSL picture is correct, the same RPA-corrected spinon machinery should transfer to other weakly ordered easy-plane cobaltates—such as the triangular-lattice Na2BaCo(PO4)2 and K2Co(SeO3)2, which the paper names as future candidates—and would predict continuum-dominated spectra there too.","(Editorial inference) The authors fix the ratio of the third-neighbor to nearest-neighbor mixing parameters ($\\alpha_3/\\alpha_1 = 0.6$) to match DMRG; a natural extension is to determine the optimal mixing parameters from the free energy in Eq. (7), which would turn the phase diagram and spectra from fitted into predictive.","(Editorial inference) Because the calculation deliberately freezes the internal gauge-field fluctuations of the spin liquid, the decisive test is whether including those fluctuations shifts the predicted ordering wavevectors; if it does, the observed magnetic states would not be a direct fingerprint of the Dirac spin liquid."],"forward_implications":["At intermediate frustration $0.32 < J_3/J_1 < 0.37$, the pure DSL and the weakly ordered states (FM, zig-zag, double zig-zag, spiral) lie within 1-2% in energy, so the DSL is the natural parent and the observed order is a weak instability of it.","The sharp modes seen in neutron scattering along $\\Gamma\\to M$ and $\\Gamma\\to K$, coexisting with a continuum at the zone center, emerge from RPA-corrected spinon dynamics, so they do not require strong magnon interactions to be explained.","The easy polarization of the ordered moments in an in-plane field (fields of order $0.5$ T given $J_1\\approx 7$ meV and $g\\approx 3$) follows because the Zeeman field gaps the Dirac nodes and flattens the $\\Gamma\\to M$ mode, matching the field-dependent neutron response.","A low-temperature $T^3$ power law in the NMR relaxation rate $1/T_1$ is a direct Dirac-spinon signature that the calculation predicts for honeycomb cobaltates."],"supporting_citations":[{"why":"Supplies the DSL ansatz and prior VMC evidence that the same honeycomb J1-J3 model hosts a proximate Dirac spin liquid.","marker":"[60]"},{"why":"Experimental neutron scattering and magnetization data on BaCo2(AsO4)2 against which the RPA-corrected spectra are compared, including in-plane field response.","marker":"[14]"},{"why":"THz spectroscopy data showing a magnetic continuum in BaCo2(AsO4)2, the low-energy peak of which the calculation reproduces at the zone center.","marker":"[41]"},{"why":"Ab initio derivation of the J1-J3 easy-plane XXZ exchange parameters and g-factor used to set energy scales J1 ≈ 7-8 meV.","marker":"[52]"},{"why":"DMRG study identifying weak magnetic order and competing phases in the intermediate frustration regime, providing the phase-diagram benchmark.","marker":"[55]"},{"why":"DMRG and experimental study of quantum fluctuations suppressing critical fields in BaCo2(AsO4)2, giving the benchmark for weak-field polarization and zig-zag/spiral competition.","marker":"[59]"},{"why":"Theoretical result that an algebraic (Dirac) spin liquid acts as the mother of many competing orders, the conceptual basis for treating the DSL as parent state.","marker":"[27]"}],"fun_headline_variants":["Dirac spin liquid parent state explains cobaltate orders and dynamics","Easy-plane Dirac spin liquid is parent of cobaltate magnetic orders","Gutzwiller RPA ties Dirac spin liquid to cobaltate spin dynamics","Cobaltate orders emerge from Dirac spin liquid parent"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole calculation rests on a tunable knob that mixes a classical spin picture with an atomic-scale quantum spin-liquid picture while freezing the internal fluctuations of the quantum picture; if those frozen fluctuations are actually important, the predicted phases and spectra are not reliable, a limitation the paper acknowledges by saying the calculation is not controlled (Sec. I).","fun_headline_variants_meta":{"raw":{"variants":["Dirac spin liquid parent state explains cobaltate orders and dynamics","Easy-plane Dirac spin liquid is parent of cobaltate magnetic orders","Gutzwiller RPA ties Dirac spin liquid to cobaltate spin dynamics","Cobaltate orders emerge from Dirac spin liquid parent"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000491,"raw_usage":{"total_tokens":2485,"prompt_tokens":1088,"completion_tokens":1397,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":704,"completion_tokens_details":{"reasoning_tokens":1325}},"tokens_in":704,"tokens_out":1397,"duration_ms":10422,"temperature":1.0,"reasoning_tokens":1325,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T21:23:57.443477+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A concrete observation that would settle the claim: measure the THz absorption of BaCo2(AsO4)2 below 0.2 THz, the range the paper identifies as experimentally unexplored (Sec. V A). At $J_3/J_1 \\approx 0.35$ the RPA-corrected model predicts an enhanced low-energy peak near $\\Omega/J_1 \\approx 0.1$; a spectrum with no such peak, or with a purely magnon-like gapped onset, would rule out the proximate-DSL RPA description.","supporting_citations":[{"cited_title":"Halloran, F","cited_arxiv_id":null,"evidence_quote":"Experimental neutron scattering and magnetization data on BaCo2(AsO4)2 against which the RPA-corrected spectra are compared, including in-plane field response."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"THz spectroscopy data showing a magnetic continuum in BaCo2(AsO4)2, the low-energy peak of which the calculation reproduces at the zone center."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Ab initio derivation of the J1-J3 easy-plane XXZ exchange parameters and g-factor used to set energy scales J1 ≈ 7-8 meV."}],"review_version":1}