{"id":"c914573e-ffd1-44c9-9803-06005712aedf","arxiv_id":"2608.08738","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Using tensor networks, the authors compute the ground-state phase diagram of a two-component extended Bose-Hubbard model, find no supersolidity at experimental parameters, and predict a supersolid phase when one hopping is increased.","lead":"This paper maps the zero-temperature phases of a two-component Bose-Hubbard model for dipolar excitons and finds no supersolid phase for the current experimental parameters. It then shows that increasing one component's hopping can stabilize a supersolid phase, suggesting a concrete experimental path forward.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No-supersolid result for experimental parameters rests on D=4 iPEPS with no convergence extrapolation; this asymmetry leaves the central negative claim unsecured.","rationale":"The reader's weakest assumption correctly identifies the key asymmetry in the paper's evidence. The positive CB-SS claim in the tuned regime is supported by a 1/D extrapolation at a representative point (Fig. 6) and by 2×2 versus 4×4 unit-cell consistency (Fig. 7), so I do not see a reason to doubt the existence of that phase. The negative claim for the experimental regime, however, has no analogous convergence check. Since the abstract's first conclusion is the absence of supersolidity in the experimentally relevant platform, the standard of evidence should match the positive claim: the phase diagram in Fig. 2 and the cuts in Fig. 3 should be shown to be stable under bond-dimension growth. Without that, a narrow supersolid region could be missed near the SF pockets, and the statement that the absence is caused by interaction-dominated scales would be premature. I agree with the reader that this warrants a conditional verdict rather than rejection: the methods are standard, the model is clearly specified, and the missing piece is a specific convergence study rather than a fundamental flaw. The paper would also be strengthened by releasing the underlying data, since the data availability statement points only to 'reasonable request,' but that is secondary to the bond-dimension check.","tokens_in":9016,"tokens_out":4655,"duration_ms":50836,"concrete_test":"Recompute the experimental-regime cuts of Fig. 3(a,b) with D=6 and D=8 (or full-update iPEPS) at fixed μ2=50 over μ1=135–150 and at fixed μ2=200 over the SF-VM window, tracking ψ1, ψ2, Δ1, Δ2 and ground-state energies, then extrapolate to 1/D→0. If the SF pockets remain isolated with ψ1→finite only inside them and Δ1→0 there, and no CB-SS sector appears, the negative claim is supported. If instead ψ1 grows or Δ1 stays finite inside a pocket as 1/D→0, the no-supersolid conclusion fails. A comparison of the extrapolated pocket width against the Fig. 2 boundaries would directly settle whether the reported phase diagram is converged.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The paper's central practical message is negative: at the experimental parameters (Eq. 6), no supersolid exists, because the superfluid pockets are tiny (Fig. 2; only Δμ1≈5 in Fig. 3a) and interaction scales suppress coherence. That claim is established in Sec. III.A entirely with D=4 iPEPS and simple update, with no D=5,6,... data and no 1/D extrapolation for the experimental parameter set. The only convergence scaling reported (Fig. 6) is for one point in the tuned CB-SS phase, whose qualitative character would not change if D=4 were inaccurate. The negative claim has a different logical status: it asserts absence of a phase over a region of parameter space. At small D, iPEPS is biased by limited entanglement, and simple update can misplace narrow compressible windows; without extrapolation it is unknown whether the order parameters ψ1, ψ2, Δ1, Δ2 at the SF pockets converge to finite values, whether the SF pocket broadens, or whether a CB-SS region appears at the pocket edges. The 4×4 unit-cell energy comparison (Fig. 7) tests unit-cell compatibility, not bond-dimension convergence, at phases already identified. Thus 'no evidence for a supersolid phase' is literally true as a report of the runs performed, but the stronger interpretive claim in the abstract—that the absence is due to interaction-dominated scales—requires ruling out a D=4 artifact. This is the most load-bearing weakness because the abstract's first conclusion for the experimental platform rests on it.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the ground-state phase diagram of a two-component extended Bose-Hubbard model with dipolar excitons, using infinite projected entangled-pair states (iPEPS) with a 2x2 unit cell and simple-update optimization. For the experimentally relevant parameters (Eq. 6), the authors report checkerboard, Mott-insulating, superfluid, and vacuum phases with orbital-selective character, and they find no supersolid phase, attributing this to the strongly interaction-dominated energy scales. For a tuned parameter set with enhanced hopping of one component (t2=35), they identify an orbital-selective checkerboard supersolid (CB-SS) phase. The CB-SS phase is supported by a finite-bond-dimension (1/D) extrapolation at one representative point (Fig. 6) and by 2x2 versus 4x4 unit-cell energy comparisons (Fig. 7). The negative no-supersolid statement for the experimental regime, however, is based entirely on D=4 iPEPS data without convergence extrapolation.","tokens_in":9355,"tokens_out":3809,"duration_ms":42687,"significance":"If the results hold, the paper provides a controlled many-body calculation of the ground-state phase structure of a recently realized dipolar-exciton platform, clarifying the absence of supersolidity in the current experimental regime and proposing a concrete route toward realizing supersolidity by enhancing the hopping asymmetry. The computational evidence for the positive CB-SS claim is genuinely strengthened by the 1/D scaling analysis and the enlarged-unit-cell energy check, which is a commendable feature. The negative result for the experimental parameter set is practically important but rests on less secure numerical footing, which limits the confidence in the central practical message.","major_comments":[{"comment":"The no-supersolid conclusion for the experimental parameter regime is based solely on D=4 simple-update iPEPS calculations with no bond-dimension scaling or 1/D extrapolation at that parameter set. The claim is a statement of absence over a region of the (mu_1, mu_2) plane; at small bond dimension, iPEPS can artificially narrow compressible windows, shift phase boundaries, or suppress order parameters near the superfluid pockets. The convergence check in Fig. 6 is performed only at a single point in the tuned CB-SS phase, which is a positive existence claim and does not control the negative result. The abstract and Sec. IV state that supersolidity is absent in the experimental regime, but the numerical evidence as presented does not rule out a D=4 artifact in the narrow SF regions. The authors should add D=5 and D=6 data (with a 1/D extrapolation if feasible) for representative cuts in the experimental regime, especially at the SF pockets and their boundaries, or explicitly qualify the negative statement as preliminary.","section":"Sec. III.A, Figs. 2 and 3"},{"comment":"The reduction to hard-core bosons (n_{i alpha}=0,1) is asserted without a numerical test. While U1=1000 and U2=500 are large, the intercomponent repulsion U3=200 and the nearest-neighbor terms V2=250 are not extremely small compared to the hoppings t1=1 and t2=7; the hard-core truncation could in principle modify the width and location of the narrow superfluid pockets that are central to the no-supersolid conclusion. The authors should provide a concrete check, for example by allowing double occupancy for one species at a representative point in the SF pocket or by citing a previous comparison, to support the statement that the truncation is 'safe in practice'.","section":"Sec. II, Eqs. (4)-(6)"}],"minor_comments":[{"comment":"The statement that discontinuities in the order parameters indicate that 'the phase transitions are all first order' is not fully established; at finite bond dimension, a discontinuous change can also result from metastability or slow convergence near a continuous transition. Please clarify the criterion used.","section":"Fig. 3 caption"},{"comment":"The error bars for the transition points are quoted as ±2 if not plotted, but the procedure for locating the transitions and estimating the errors is not described in Sec. II. A brief description of the scan and refinement method would improve reproducibility.","section":"Sec. III.B, Fig. 4 caption"},{"comment":"The environment dimension chi is stated to be chosen at least D^2 and 'sufficient for converged expectation values', but no convergence test in chi is shown or described. A short sentence with a representative check would strengthen the numerics section.","section":"Sec. II, CTMRG paragraph"},{"comment":"The abstract uses the careful phrasing 'we find no evidence for a supersolid phase', but Sec. IV states 'we find no supersolid ground state'. These have different logical strengths; the conclusion should match the level of numerical certainty, particularly given the bond-dimension limitations discussed above.","section":"Abstract and Sec. IV"},{"comment":"The definition of the condensate order parameter sums the absolute values of <b_{i alpha}> over the unit cell. Please state explicitly that a uniform phase is imposed or that the absolute value is taken after gauge fixing, so that a staggered phase pattern is not misinterpreted as incoherence.","section":"Sec. II, Eq. (10)"},{"comment":"The sentence 'Guided by the conclusion of the one-dimensional counterpart [37]' is vague; please specify which conclusion (e.g., that enhanced hopping stabilizes a supersolid or supercounterfluid) motivates the choice t2=35.","section":"Sec. III.B, first paragraph"}],"recommendation":"major_revision","confidential_remarks":"The main practical message of the paper is the negative result for the experimental regime. That claim is currently supported only by D=4 iPEPS data, and the paper's own convergence analysis is restricted to the tuned CB-SS phase. The discrepancy in the weight of numerical evidence between the negative and positive claims is the central weakness. The positive CB-SS result is reasonably solid, and the manuscript is well within the scope of the journal. I would encourage the editor to request a revision that either provides convergence data for the experimental regime or explicitly softens the conclusions to match the current numerical evidence."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe paper is worth reading for one thing: it gives the first tensor-network ground-state phase diagram for the dipolar-exciton two-component EBH model, and it tells you where the supersolid is not and where it could be. The tuned t2=35 route is concrete and testable.\n\nWhat's genuinely new: the iPEPS phase diagram in the (μ1, μ2) plane for the experimental parameters, and the identification of orbital-selective phases (e.g., CB-VM). The CB-SS phase in the tuned regime is supported by a 1/D extrapolation over D=3–8 at one representative point, plus 2×2 vs 4×4 energy checks. That's real evidence and the main positive result appears robust.\n\nThe soft spot is exactly where the stress test lands: the no-supersolid conclusion for the experimental regime rests entirely on D=4 simple-update iPEPS with no convergence extrapolation at any point in that regime. The one extrapolation shown (Fig. 6) is for the CB-SS phase, which is not the regime where the paper makes its central negative claim. The abstract says absence is due to interaction-dominated scales, but to rule out a D=4 artifact you'd want to see at least D=5 or D=6 at a representative boundary point (e.g., in the SF sliver of Fig. 3a) and a 1/D trend for ψ and Δ. Without that, \"no evidence for a supersolid\" is literally what the runs show, but the stronger claim about the experimental platform being firmly localized is under-supported.\n\nAlso minor: the hard-core boson simplification is reasonable given U1=1000, U2=500, but a sentence justifying the truncation to n=0,1 against the full model would help. Data are only 'available upon request'—fine, but not ideal.\n\nThe citation pattern looks fair; Ref. [23] is the experimental source and Ref. [37] (overlapping authors) is used only to motivate the t2 increase, which is a parameter scan, not a fitted result. No circularity.\n\nBottom line: the paper deserves a serious referee. It is a potentially useful phase diagram with a concrete prediction. The referee should ask for convergence checks on the experimental-regime negative result, or for the abstract to be softened to 'no evidence at D=4.' I'd send it to review.","headline":"Solid iPEPS phase diagram with a testable supersolid route, but the no-supersolid claim for the experimental regime needs a convergence check before you trust it.","tokens_in":9865,"tokens_out":2316,"would_cite":true,"duration_ms":24736,"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":"Using tensor-network ground-state simulations, this paper establishes that the experimentally realized two-component extended Bose–Hubbard model has no supersolid ground state, and shows that increasing the hopping of one component to…","keywords":["extended Bose-Hubbard model","two-component bosons","dipolar excitons","supersolid","orbital-selective phases","tensor network","infinite projected entangled-pair states","ground-state phase diagram"],"falsifier":"Run the same iPEPS calculation at $D=6$ or $D=8$ along the narrow superfluid corridor near $\\mu_1=140$–$145$ at $\\mu_2=50$ and extrapolate $1/D$ to the thermodynamic limit; if the density-modulation order parameter $\\Delta_1$ stays strictly zero wherever $\\psi_1>0$, the no-supersolid claim survives, whereas a finite $\\Delta_1$ in the converged limit would overturn it.","tokens_in":8861,"feed_emoji":"🧊","tokens_out":13352,"duration_ms":115091,"temperature":0.7,"pith_summary":"This paper asks whether the two-component extended Bose–Hubbard model realized with dipolar excitons can host a supersolid ground state, and answers it with a numerical phase diagram in the plane of the two chemical potentials. For the parameter values extracted from the experiment ($t_1=1$, $t_2=7$, $U_1=1000$, $U_2=500$, $U_3=200$, $V_1=35$, $V_2=250$, $V_{12}=V_{21}=20$), the ground state is dominated by checkerboard, Mott-insulating, and vacuum phases, with only tiny superfluid regions and no coexistence of density order with phase coherence. The authors attribute this absence to interaction-dominated energy scales that suppress superfluidity. They then show that raising one hopping parameter to $t_2=35$ moves the system into a nearby regime where an orbital-selective checkerboard supersolid phase appears and survives bond-dimension and unit-cell checks. If correct, the paper pins down the zero-temperature structure of the dipolar-exciton platform and gives experiments a concrete target parameter for supersolidity.","feed_headline":"No supersolid in dipolar-exciton lattice; raising t2 to 35 creates one","feed_subtitle":"Tensor-network maps put the experimental regime in localized phases and chart a route to a supersolid.","key_machinery":"The load-bearing object is the two-component extended Bose–Hubbard Hamiltonian with dominant on-site terms $U_1 n_{i1}^2 + U_2 n_{i2}^2 + U_3 n_{i1} n_{i2}$ and nearest-neighbor density-density interactions $V_1$, $V_2$, $V_{12}$, $V_{21}$ between the two orbitals, together with the hard-core constraint that each site holds at most one boson per component. The method that carries the argument is the infinite projected entangled-pair state (iPEPS) tensor network on a square lattice, optimized by simple-update imaginary-time evolution with 2×2 (and, for checks, 4×4) unit cells. Phase classification is done through three order parameters: average density $\\rho_\\alpha$, checkerboard density modulation $\\Delta_\\alpha$, and condensate amplitude $\\psi_\\alpha$. A supersolid is identified specifically as a state with both $\\psi_\\alpha > 0$ and $\\Delta_\\alpha > 0$ in the same component, and the paper's key discriminations—absence of supersolidity in the experimental regime and its presence after tuning—are made by computing these quantities as functions of $\\mu_1$ at fixed $\\mu_2$.","core_discovery":"The central claim is that the experimental regime of this two-component extended Bose–Hubbard model has no supersolid ground state, while a modest increase of the second component's hopping to $t_2 = 35$ stabilizes an orbital-selective supersolid. In the experimental regime, the tensor-network phase diagram contains only VM-CB, CB-VM, CB-CB, MI-VM, VM-SF, and SF-VM sectors; the two superfluid sectors occupy a tiny fraction of the $(\\mu_1, \\mu_2)$ plane, and in them the density-modulation order parameter $\\Delta_\\alpha$ vanishes while the condensate order parameter $\\psi_\\alpha$ is finite, so there is no window in which both orders coexist. The proposed reason is the large on-site and nearest-neighbor interaction scales relative to hopping, which localize the bosons and restrict superfluidity. With $t_2$ increased from 7 to 35, the former CB-VM region is replaced by a CB-SS phase in which component 1 keeps checkerboard order while component 2 simultaneously has $\\psi_2 > 0$ and $\\Delta_2 > 0$; a $1/D$ extrapolation gives $\\psi_2 = 0.260(6)$, $\\rho_2 = 0.1396(16)$, $\\Delta_2 = 0.0714(20)$, and 2×2 and 4×4 unit cells give nearly identical energies. The authors conclude that the hopping asymmetry is the control knob for supersolidity in this platform.","pith_inferences":["Beyond the paper, if the $D=4$ result for the experimental regime is not fully converged, the small superfluid pockets could widen at larger bond dimension, so a $1/D$ study of the negative result—not just of the tuned supersolid—is the natural next check.","Beyond the paper, the success of the $t_2=35$ route suggests that other parameter paths, such as reducing the nearest-neighbor repulsion $V_2$ or the intercomponent repulsion $U_3$ while keeping $t_2$ moderate, may stabilize supersolidity at even lower hopping ratios; the paper does not explore these.","Beyond the paper, at finite temperature the orbital-selective supersolid should melt through a sequence of transitions—phase coherence lost before density order, or vice versa—so experiments searching for it will need to specify temperature relative to the first-order boundaries mapped here.","Beyond the paper, the same two-orbital hard-core structure appears in other artificial-lattice platforms such as Rydberg-dressed or dipolar atomic gases, so the orbital-selective checkerboard-supersolid mechanism may transfer to those settings if the hopping asymmetry can be engineered."],"forward_implications":["The experimentally realized parameter set should be treated as strongly localized at zero temperature, so observations of superfluidity or supersolidity in that exact regime would require physics beyond the model as parameterized.","Raising the hopping $t_2$ to 35 while keeping the other parameters fixed is a concrete route: the CB-SS phase should appear around $\\mu_1 \\approx 75$ at $\\mu_2 = 80$ and persist over a finite region of the chemical-potential plane.","Phase transitions in both regimes are first order, signaled by discontinuities in the order parameters, so tuning through the phase boundaries will encounter hysteresis or phase coexistence rather than continuous critical behavior.","Orbital-selective order is generic in this model: one component can sit in a checkerboard or Mott state while the other is superfluid, vacuum, or supersolid, so phase diagrams must be labeled per component.","The absence of supersolidity in the experimental regime is not a failure of the model but a statement about energy scales: interaction-dominated parameters suppress the required coexistence of density order and phase coherence."],"supporting_citations":[{"why":"Supplies the experimental platform, the two Wannier orbitals, and the full parameter set whose phase diagram is the paper's primary object.","marker":"[23]"},{"why":"Introduces the simple-update iPEPS optimization scheme used to obtain all ground states.","marker":"[27]"},{"why":"Provides the tensor-network formalism and notation for the iPEPS ansatz and its virtual bond dimension.","marker":"[26]"},{"why":"Supplies the corner transfer-matrix renormalization group methods used to contract the tensor network and evaluate observables.","marker":"[32–35]"},{"why":"Gives the single-component hard-core boson phase diagram the paper contrasts with its two-component results.","marker":"[36]"},{"why":"Guides the tuning strategy by showing how enhanced hopping stabilizes coherence in the one-dimensional two-component system.","marker":"[37]"},{"why":"Supplies the concept of orbital-selective behavior used to categorize phases in which the two components occupy different quantum states.","marker":"[24]"}],"fun_headline_variants":["No supersolid in dipolar-exciton lattice—until t2 jumps to 35","Dipolar-exciton model: no supersolid in experiment, but hopping boost to t2=35 works","Ground-state map shows supersolid absent in dipolar excitons; route via t2=35","Supersolid elusive in dipolar-exciton lattice; raising t2 reveals it"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The negative result for the experimental regime is computed at bond dimension $D=4$ with no $1/D$ extrapolation, so the conclusion that no supersolid exists there assumes that $D=4$ already captures the narrow superfluid pockets faithfully.","fun_headline_variants_meta":{"raw":{"variants":["No supersolid in dipolar-exciton lattice—until t2 jumps to 35","Dipolar-exciton model: no supersolid in experiment, but hopping boost to t2=35 works","Ground-state map shows supersolid absent in dipolar excitons; route via t2=35","Supersolid elusive in dipolar-exciton lattice; raising t2 reveals it"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000416,"raw_usage":{"total_tokens":2192,"prompt_tokens":1035,"completion_tokens":1157,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":651,"completion_tokens_details":{"reasoning_tokens":1058}},"tokens_in":651,"tokens_out":1157,"duration_ms":9543,"temperature":1.0,"reasoning_tokens":1058,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T04:25:34.918658+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the same iPEPS calculation at $D=6$ or $D=8$ along the narrow superfluid corridor near $\\mu_1=140$–$145$ at $\\mu_2=50$ and extrapolate $1/D$ to the thermodynamic limit; if the density-modulation order parameter $\\Delta_1$ stays strictly zero wherever $\\psi_1>0$, the no-supersolid claim survives, whereas a finite $\\Delta_1$ in the converged limit would overturn it.","supporting_citations":[{"cited_title":"Altman, W","cited_arxiv_id":null,"evidence_quote":"Supplies the experimental platform, the two Wannier orbitals, and the full parameter set whose phase diagram is the paper's primary object."},{"cited_title":"H´ ebert, G","cited_arxiv_id":null,"evidence_quote":"Guides the tuning strategy by showing how enhanced hopping stabilizes coherence in the one-dimensional two-component system."},{"cited_title":"Lagoin, U","cited_arxiv_id":null,"evidence_quote":"Supplies the concept of orbital-selective behavior used to categorize phases in which the two components occupy different quantum states."}],"review_version":1}