{"id":"941b0140-281c-4ad8-9a5b-f701e5090537","arxiv_id":"2502.00542","paper_version":2,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"low","formal_verification":"none","parameter_count":1,"one_line_summary":"The drag coefficient in the dilute-hole limit of the hard-core two-component Bose-Hubbard model on the square lattice is exactly 1-2/π, about 0.36, and the effect is carried by hole-spin polarons.","lead":"This paper derives an exact formula for the superfluid drag between two bosonic components in a hard-core lattice model, finding a large drag coefficient of 1-2/π at low hole density. The result gives a benchmark for strongly correlated lattice superfluids and points to an observable spin-correlation signature around holes.","discovery_kind":"first_principles","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified: the single-spin-flip truncation is controlled by a selection rule, and the finite-density extrapolation does not affect the exact x_v→0 claim.","rationale":"The central claim is the exact O(q^2) single-hole energy in Eq. (21). I examined the only plausible failure point: the truncation to one flipped spin in Eq. (7). This truncation is not an uncontrolled approximation. The spin-flip coupling in Eq. (6) is proportional to sin(q·r), and each application changes the number of flipped spins by exactly ±1. Therefore V|ψ0⟩ is a one-flip state, the O(q^2) energy shift is exactly the standard second-order expression with one-flip intermediates, and two-flip sectors enter only at O(q^4). Equations (12)–(13) are thus the complete O(q^2) Schrödinger equation, and the integral (22) follows. The q=0 reference state is the fully polarized XY ferromagnet; for a single hole this is the Nagaoka ground state, so the derivation does not rely on a fragile ferromagnetic assumption. The only genuine numerical soft spot is the C→∞ extrapolation from C=3,4,5 using a 1/C^2 scaling and the QMC limitation at small x_v, but these do not affect the analytic x_v→0 limit, whose finite-C values are independently matched by the iMPS stars in Figs. 1 and 2. The reader's ACCEPT with high confidence remains appropriate.","tokens_in":14226,"tokens_out":23890,"duration_ms":253004,"concrete_test":"Perform exact diagonalization of the hard-core two-component Bose-Hubbard model on a 4×4 torus in the single-hole sector with twisted boundary conditions q, keeping the full multi-spin-flip Hilbert space, and extract the coefficient of q^2 in the ground-state energy. Compare with the finite-size version of Eq. (22) obtained by replacing the integral with sums over kx,ky; if the full Hilbert-space curvature disagrees with the one-spin-flip result beyond numerical precision, the central claim is invalidated.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The reader's flagged weakest assumption—that one-spin-flip states suffice at O(q^2)—is actually exact to that order. The spin-flip perturbation in Eq. (6) is linear in q and changes the number of flipped spins by exactly ±1, so its action on the fully polarized state produces a one-flip intermediate state. The O(q^2) energy shift is therefore exactly the second-order term with one-flip intermediates; multi-flip amplitudes are generated at O(q^2) in the wavefunction and contribute to the energy only at O(q^4). Equations (12)–(13) are thus the complete O(q^2) Schrödinger equation, and Eq. (21) with κ from Eq. (22) follows without an uncontrolled variational truncation. The q=0 reference state is the fully polarized XY ferromagnet; for a single hole this is the Nagaoka ground state, not a separate assumption. The only genuine numerical soft spot is the C→∞ extrapolation using ρ_−(C)=ρ_−∞+α/C^2 with only C=3,4,5, and the QMC limitation at small x_v, but these do not bear on the analytic limit, which is independently supported by the finite-C analytic values matching the iMPS stars. No load-bearing concern with the central claim survives scrutiny.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper studies superfluid drag in the two-component hard-core Bose-Hubbard model in the limit of infinite on-site repulsion. On the square lattice, the authors derive an exact expression for the drag coefficient in the vanishing-hole-density limit, κ = 1 − 2/π (Eqs. 21–23), by solving the single-hole Schrödinger equation in the presence of a phase twist to second order in the twist. They also obtain exact values for finite cylinders and other lattices. The analytic result is compared with two independent numerical methods: infinite cylinder iMPS (VUMPS) calculations and quantum Monte Carlo on finite tori, finding good agreement. The paper further characterizes the spin-hole correlations (spin polaron) and discusses experimental realization in cold atoms, Rydberg arrays, and transmon arrays.","tokens_in":14485,"tokens_out":11950,"duration_ms":107382,"significance":"The central analytic result is a rare exact, parameter-free prediction for a strongly correlated two-component lattice superfluid: the drag coefficient is fixed by the Schrödinger equation without fitting, and the same calculation yields explicit finite-cylinder values that match the numerical data. The paper's convergence of three methods—analytic perturbation theory, iMPS, and QMC—gives strong support to the claim. The result is falsifiable and has clear experimental relevance through the spin-hole correlator Υ_j. The work also provides a useful general picture: in the dilute-hole limit, the spin channel stiffness is strongly reduced by correlated-hopping processes, giving an order-unity drag coefficient.","major_comments":[],"minor_comments":[{"comment":"In the definition of Π^y_s, the second term should read Λ_{s_x, s_y − 1} rather than Λ_{s_x, s_y + 1}; the following equations use the antisymmetric combination, so this is a typographical error.","section":"Appendix A, Eq. (A3)"},{"comment":"The intermediate expression E = ϵ_0 − 2q^2t Π^x_x/(1 − Π^x_x) should have a plus sign in the denominator (1 + Π^x_x); the subsequent expression for E/t in Eq. (A8) is consistent with the plus sign, so the minus sign is a typo.","section":"Appendix A, text after Eq. (A7)"},{"comment":"The phrase 'fs,0 ∝ q is small' is misleading: the no-flip amplitude f0 is order one (f0 = 1 + O(q^2)), while the flip amplitudes f_s with s≠0 are O(q). Please reword to avoid confusion.","section":"Sec. V, around Eq. (7)"},{"comment":"The reported values for C = 3, 4, 5 should be written with parentheses to avoid ambiguity, e.g., κ = 2 − √(7/3) for C = 3; without parentheses, 2 − √7/3 could be read as 2 − (√7)/3.","section":"Sec. V, finite-cylinder results"},{"comment":"The C→∞ extrapolation of ρ_−(C) rests on only three cylinder widths (C = 3, 4, 5) and an assumed 1/C^2 scaling; an estimate of the extrapolation uncertainty would be helpful, although the central analytic claim does not rely on this extrapolation.","section":"Sec. VIII and Appendix B"},{"comment":"The statement that κ_nD ∼ 1/(z−1) on hypercubic lattices is given without derivation; a brief derivation or a more explicit statement of the asymptotic regime would improve the presentation.","section":"Sec. V, generalization to other lattices"}],"recommendation":"minor_revision","confidential_remarks":"The manuscript presents a strong, well-supported exact result for the superfluid drag coefficient. The central derivation is sound; the issues are limited to typographical and presentational points. I recommend minor revision. The paper is a good fit for the journal's scope."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The headline result is real: in the vanishing-hole-density limit of the hard-core two-component Bose-Hubbard model on the square lattice, the superfluid drag coefficient is exactly κ = 1 − 2/π. That is a clean, parameter-free statement about a canonical strongly correlated lattice model, and it is not in the prior literature. The polaron interpretation—holes dressed by flipped spins in the presence of counterflow—is also new and physically illuminating. On top of that, the paper gives exact finite-cylinder and other-lattice values, which is a nice bonus.\n\nWhat the paper does well: the analytic derivation in Sec. V and Appendix A is coherent and actually controlled. The one-spin-flip truncation looks like an assumption, but the spin-flip term in Eq. (6) changes the number of flipped spins by exactly ±1, so the O(q^2) energy shift only needs one-flip intermediate states. Multi-flip contributions to the wavefunction enter at O(q^2) but affect the energy only at O(q^4). So the closed form for κ is not a variational guess; it is the exact O(q^2) result. The two numerical methods, iMPS on cylinders and worm-algorithm QMC on tori, agree with each other and with the analytic limit. The paper is also transparent about where the numerics are weakest, especially in Appendix B.\n\nSoft spots, in proportion: the finite-density extrapolation ρ_−(C) = ρ_−∞ + α/C^2 uses only C = 3, 4, 5, which is thin, and the QMC data have limited reach at small x_v due to kinetic constraints and autocorrelation times. But these are clearly flagged, and they do not bear on the exact dilute-hole claim, which is independently supported by the finite-C analytic values matching the iMPS stars. The ferromagnetic ground-state assumption for the single hole is invoked from Nagaoka-type physics rather than proven from the Hamiltonian, but it is standard and consistent with all numerics.\n\nWho this is for: quantum-gas theorists who care about Andreev–Bashkin drag in lattices, and cold-atom experimentalists looking for a measurable strong-drag signature. The spin-hole correlator Υ_j is a concrete observable. This paper deserves a serious referee; I would send it to peer review with confidence, expecting only minor revisions around the extrapolation discussion.","headline":"Exact single-hole drag coefficient κ = 1 − 2/π in the hard-core two-component Bose-Hubbard model, backed by two independent numerics; the paper is solid and deserves serious refereeing.","tokens_in":15051,"tokens_out":1402,"would_cite":true,"duration_ms":15840,"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":"Single hole fixes superfluid drag at exactly 1−2/π","keywords":["superfluid drag","Andreev-Bashkin effect","two-component Bose-Hubbard model","hard-core bosons","kinetic magnetism","spin polaron","quantum Monte Carlo","matrix product states"],"falsifier":"A high-accuracy calculation of the energy change under a small counterflow twist at very low hole density on a two-dimensional square lattice would settle the claim: if the drag coefficient extrapolated to zero hole density disagrees with 1 − 2/π beyond finite-size error, the one-spin-flip ansatz is wrong. Alternatively, a quantum-gas-microscope measurement of the spin twist around a hole that does not show the predicted long-distance tail Υ_j ∝ (j_x q_x + j_y q_y)/|j|² would cast doubt on the mechanism.","tokens_in":14001,"feed_emoji":"⚛️","tokens_out":10016,"duration_ms":83351,"temperature":0.7,"pith_summary":"This paper establishes an exact value for the superfluid drag coefficient in the two-component Bose-Hubbard model with infinitely strong repulsion on the square lattice. In the limit of vanishing hole density, a single mobile hole produces a strong dissipationless coupling between the currents of the two components, and the drag coefficient is exactly κ = 1 − 2/π. The authors derive this closed form by solving the single-hole Schrödinger equation with a variational ansatz that allows one flipped spin, and they verify it with infinite matrix product state calculations on cylinders and quantum Monte Carlo on finite tori. The result turns a strongly correlated many-body response into a solvable single-hole problem, and it predicts a measurable spin polaron around each hole.","feed_headline":"Single hole fixes superfluid drag at 1−2/π","feed_subtitle":"At vanishing hole density, a single mobile hole pins the drag coefficient to exactly 1−2/π.","key_machinery":"The central object is the single-hole variational wavefunction |ψ⟩ = (1/√N_s) ∑_j b_{j+} [ f_0 + ∑_s f_s b†_{j+s,−} b_{j+s,+} ] |ϕ⟩, which keeps exactly one spin flip relative to the fully polarized XY ferromagnet. Solving the resulting Schrödinger equations for the amplitudes f_0 and f_s reduces the drag coefficient to a Brillouin-zone integral, κ = ∫ d²p/(2π)² (cos 2p_x − 1)/(−4 + 2 cos p_x + 2 cos p_y), which evaluates to 1 − 2/π. The mechanism that carries the argument is correlated hopping: a flipped spin adjacent to the hole can hop over it, and this correlated process couples the density and spin channels, producing the drag.","core_discovery":"On its own terms, the paper claims that the superfluid density matrix of the hard-core two-component Bose-Hubbard model on a square lattice has eigenvalues ρ_+ = x_v and ρ_− = x_v/(π − 1) in the limit x_v → 0, giving a drag coefficient κ = (ρ_+ − ρ_−)/(ρ_+ + ρ_−) = 1 − 2/π ≈ 0.363. The argument is exact for a single hole: after a gauge transformation that imposes phase twists, the current-carrying ground state is an XY ferromagnet dressed by one flipped spin, and the correlated hopping of that spin with the hole depletes the spin-channel stiffness. The same calculation gives κ_1D = 1 on a chain, κ ≈ 0.23 on the triangular lattice, and κ ≈ 1/(z − 1) on hypercubic lattices, so the drag weakens as the coordination number grows. The authors verify the square-lattice value against tensor-network and Monte Carlo simulations at finite hole density, finding agreement in the extrapolated zero-doping limit.","pith_inferences":["The single-hole solution could serve as the zeroth order of a controlled expansion in hole density, giving a systematic analytic route to finite-density drag beyond numerics.","An exact two-dimensional transport coefficient of this kind is rare, so the closed form provides a benchmark against which tensor-network and quantum Monte Carlo codes for strongly correlated lattice models can be tested.","Measuring the hole-spin correlator in a quantum gas microscope could provide a model-independent test of the drag mechanism, since the predicted polaron shape encodes the same physics as κ."],"forward_implications":["In the dilute-hole limit, counterflow currents are strongly suppressed relative to mass currents, so moving one component inevitably drags the other along.","The spin-channel superfluid density at low doping is ρ_− = x_v/(π − 1), a sharp quantitative prediction that deviates from the mean-field expectation ρ_+ = x_v.","The drag coefficient sets the magnitude and sign of vortex–vortex interactions between the two species, so a drag of order 0.36 should produce readily observable vortex coupling.","The single-hole calculation is exact for any regular lattice, yielding testable predictions for chains, hypercubic lattices, and the triangular lattice.","The spin polaron around a hole, with twist decaying as Υ_j ∝ (j_x q_x + j_y q_y)/|j|² at long distances, is observable with quantum gas microscopes."],"supporting_citations":[{"why":"Establishes the concept of superfluid drag as dissipationless coupling between superflows, the phenomenon under study.","marker":"[1]"},{"why":"Prior Monte Carlo study of drag in this class of models, which the paper's numerics improve and compare against.","marker":"[7]"},{"why":"Source of the kinetic ferromagnetism that fixes the single-hole ground state as fully polarized.","marker":"[16]"},{"why":"Extends the symmetric-combination ground state argument that underpins the variational ansatz.","marker":"[17]"},{"why":"Provides the uniform-MPS technique for extracting superfluid density from energy curvature.","marker":"[32]"},{"why":"Supplies the VUMPS algorithm used to optimize the cylinder ground states.","marker":"[35]"},{"why":"The finite-entanglement scaling procedure used to extrapolate tensor-network results to infinite bond dimension.","marker":"[37]"},{"why":"The original worm algorithm that the Monte Carlo simulations are based on.","marker":"[38]"},{"why":"Companion worm-algorithm formulation for winding-number statistics used to extract stiffness.","marker":"[39]"},{"why":"Adapts the worm algorithm to two-component hard-core bosons, the starting point for the QMC here.","marker":"[40]"}],"fun_headline_variants":["Single hole pins superfluid drag to 1−2/π","One vacancy fixes drag coefficient at 1−2/π","Exact drag from a single hole in hard-core bosons","A hole sets drag: 1−2/π on square lattice","Vacancy-assisted drag becomes exact at zero doping"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The derivation rests on the assumption that the wavefunction needs at most one flipped spin; if a second flipped spin changes the energy at the same order as the counterflow twist, the closed-form value of the drag coefficient would be different.","fun_headline_variants_meta":{"raw":{"variants":["Single hole pins superfluid drag to 1−2/π","One vacancy fixes drag coefficient at 1−2/π","Exact drag from a single hole in hard-core bosons","A hole sets drag: 1−2/π on square lattice","Vacancy-assisted drag becomes exact at zero doping"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000458,"raw_usage":{"total_tokens":2265,"prompt_tokens":879,"completion_tokens":1386,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":495,"completion_tokens_details":{"reasoning_tokens":1300}},"tokens_in":495,"tokens_out":1386,"duration_ms":13020,"temperature":1.0,"reasoning_tokens":1300,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-09T18:37:14.384850+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A high-accuracy calculation of the energy change under a small counterflow twist at very low hole density on a two-dimensional square lattice would settle the claim: if the drag coefficient extrapolated to zero hole density disagrees with 1 − 2/π beyond finite-size error, the one-spin-flip ansatz is wrong. Alternatively, a quantum-gas-microscope measurement of the spin twist around a hole that does not show the predicted long-distance tail Υ_j ∝ (j_x q_x + j_y q_y)/|j|² would cast doubt on the mechanism.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes the concept of superfluid drag as dissipationless coupling between superflows, the phenomenon under study."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Prior Monte Carlo study of drag in this class of models, which the paper's numerics improve and compare against."},{"cited_title":"Eisenberg and E","cited_arxiv_id":null,"evidence_quote":"Extends the symmetric-combination ground state argument that underpins the variational ansatz."},{"cited_title":"Zauner-Stauber, L","cited_arxiv_id":null,"evidence_quote":"Supplies the VUMPS algorithm used to optimize the cylinder ground states."},{"cited_title":"Prokof’ev, B","cited_arxiv_id":null,"evidence_quote":"The original worm algorithm that the Monte Carlo simulations are based on."},{"cited_title":"Prokof’ev, B","cited_arxiv_id":null,"evidence_quote":"Companion worm-algorithm formulation for winding-number statistics used to extract stiffness."},{"cited_title":"Capogrosso-Sansone, G","cited_arxiv_id":null,"evidence_quote":"Adapts the worm algorithm to two-component hard-core bosons, the starting point for the QMC here."}],"review_version":1}