{"id":"4ea28c1c-ce4f-4999-8986-f7cebc07667f","arxiv_id":"2507.16446","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"A sublattice Stokes polarimeter reconstructs Bloch wavefunctions, eigenenergies, and the quantum geometric tensor in bipartite photonic lattices from four phase-controlled photoluminescence measurements.","lead":"Researchers built an optical polarimeter that works on the two sublattices of honeycomb microcavity arrays, not on light polarization, and used it to reconstruct the full wavefunction of lattice modes across momentum space. This gives access to the Hamiltonian and geometric properties of photonic bands, including at points where bands touch.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The quantum-metric extraction in Methods E is quantitatively wrong as written: (1−Re⟨u_k|u_{k+dk}⟩)/dk² equals (g+a²)/2, not the quantum metric g, so the published QGT map and trace in Fig. 5f are not established.","rationale":"After reading in good faith, the sublattice Stokes inversion, the four intensity combinations, and the eigenstate reconstruction are internally consistent, and the SI mask analysis is honest about the point-like assumption. The most load-bearing weakness is the quantum-metric formula in Methods E: it is not merely under-specified, it is wrong for a generic gauge-fixed state. Because the abstract and conclusion advertise the quantum geometric tensor as a central outcome, the paper cannot be accepted as is; it requires correcting the formula and re-deriving the measured metric. The reader's condition on Methods E identified the region but not the algebraic error, so agreement is partial. Verdict stays CONDITIONAL, with the condition sharpened to a corrected metric extraction and revalidation.","tokens_in":20175,"tokens_out":12388,"duration_ms":138878,"concrete_test":"Recompute the quantum metric from the same reconstructed eigenstates using the gauge-invariant finite-difference expression g_xx = [1−|⟨u_k|u_{k+dk}⟩|²]/dk_x² (equivalently Re⟨∂_x u|(1−|u⟩⟨u|)|∂_x u⟩), and compare with Fig. 5f and Extended Data Fig. 8. If the trace or components change by more than ~10% or no longer match the tight-binding insets, the QGT claim fails as written. Minimal analytical check: evaluate both formulas on |u(k)⟩=(√3/2, (1/2)e^{ikx})ᵀ, where the exact metric is 3/16 but Methods E gives 1/8.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The paper's central claim includes 'full access to the system Bloch Hamiltonian and quantum geometric tensor.' The Hamiltonian and eigenstate reconstruction are carefully supported, but the quantum metric is not: the Methods E formula is algebraically incorrect. For a normalized band |u_k⟩, expanding the overlap gives ⟨u_k|u_{k+dk}⟩ = 1 + i a_x dk − (g_xx + a_x²)dk²/2 + O(dk³), with a_x = Im⟨u|∂_x u⟩ and g_xx = Re⟨∂_x u|(1−|u⟩⟨u|)|∂_x u⟩. Thus Re(1−⟨u_k|u_{k+dk}⟩) = (g_xx + a_x²)dk²/2, so g_xx = (1−⟨u1|u2⟩)/dk_x² as written is complex, gauge-dependent, and yields (g+a²)/2 rather than g. Concretely, for |u(k)⟩ = (√3/2, (1/2)e^{ikx})ᵀ, the exact metric is 3/16 while Methods E gives 1/8. The same defect affects g_yy and g_xy. Since Fig. 5f and Extended Data Fig. 8 are presented as measured quantum-geometric quantities, and the insets claim agreement with tight-binding simulations, the metric part of the central claim is unsupported as written unless the simulations used the same erroneous formula (not stated). The rest of the paper is not invalidated by this, but the QGT claim needs a corrected, gauge-invariant extraction.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper introduces a sublattice Stokes polarimeter for bipartite photonic lattices. It uses four momentum-resolved photoluminescence intensity measurements with SLM-imposed phase and amplitude masks on the A and B sublattices (IA+B, IA+iB, IA, IB) and combines them in Eq. (3) to obtain maps I0,...,I3 that isolate the Stokes components of each Bloch mode. The authors apply the protocol to staggered and gapless honeycomb micropillar lattices, reconstruct the Bloch eigenstates (theta, phi) across the Brillouin zone, extract the Hamiltonian Pauli coefficients lambda_i(k) including a next-nearest-neighbor term, and compute Berry curvature, valley Chern numbers, and the quantum metric. In the gapless case, a density-matrix formulation is used to separate overlapping bands and track eigenstates through the Dirac points with sub-linewidth precision.","tokens_in":20523,"tokens_out":11577,"duration_ms":132898,"significance":"If the issues below are resolved, this is a significant methodological advance: it generalizes Stokes polarimetry from optical polarization to the sublattice pseudo-spin and gives direct, model-independent access to Bloch eigenstates in photonic lattices. The intensity algebra in Eq. (2)-(3) follows cleanly from the stated Mueller matrices; the reconstructed phase windings, the eigenstate orthogonality check, and the valley Chern numbers provide strong evidence that the method works; and the density-matrix treatment of overlapping bands at the Dirac point is an elegant route to sub-linewidth band separation. The paper also ships explicit simulations of the mask-discretization assumption and its failure modes in the Supplementary Information. The quantum-metric extraction, however, is not correct as written and currently prevents the paper from supporting the advertised full access to the quantum geometric tensor.","major_comments":[{"comment":"As written, these expressions are not the quantum metric. For a normalized band, expanding the overlap gives Re(1 - <u_k|u_{k+dk}>) = (g_xx + a_x^2) dk^2 / 2, where a_x = Im<u|d_x u> is the Berry connection, and the overlap itself is complex unless a_x = 0. The manuscript's g_xx = (1 - <u1|u2>)/dk_x^2 is therefore complex, gauge-dependent, and even after taking the real part it returns (g_xx + a_x^2)/2 rather than g_xx. Since Fig. 5f and Extended Data Fig. 8 are presented as measured quantum-metric maps and the insets claim agreement with tight-binding simulations, the QGT part of the central claim is unsupported as written. The fix is straightforward - use the real part of the overlap with a parallel-transport gauge (or subtract a_x^2) and include the factor 1/2 - but the figures and text must be recomputed. The Berry-curvature and valley-Chern results are not affected because they use the gauge-invariant Fukui-Hatsugai-Suzuki plaquette phase.","section":"Methods E, formulas for g_xx, g_yy, g_xy"},{"comment":"The reconstruction of lambda_i(k) from the measured eigenstates is not specified by any equation, and the parameters used in the tight-binding comparison (t, epsilon0, t') are stated without indicating which are fixed by fabrication, which are extracted from the same measured dispersion, and which are fitted to the lambda0 map. Since Fig. 5 is presented as a reconstruction of the Hamiltonian and Extended Data Fig. 7 uses the lambda0 modulation to infer t', the reader cannot reproduce the central Hamiltonian claim without this information. Please add the explicit reconstruction formula (e.g., form H from the measured E_n and |u_n>, then lambda_i = Tr[H sigma_i]/2) and state the parameter provenance and uncertainties.","section":"Main text, Hamiltonian reconstruction before Fig. 5; Extended Data Fig. 7"}],"minor_comments":[{"comment":"The denominator of the S_nu,n,k formula should sum over mu of I_mu, not I_nu; as printed, the expression would collapse to sign(I_nu)/sqrt(3) and cannot produce the Stokes maps shown.","section":"Methods D, S_nu formula"},{"comment":"There are typos: 'Recontructed' should be 'Reconstructed' in the Fig. 4 caption, and 'Extanded Data Fig.' appears twice in Methods B.","section":"Fig. 4 caption; Methods B"},{"comment":"The slowly-varying-Bloch-phase assumption |k| R_m << 1 is explicitly acknowledged in the SI; the main text should state this validity condition and refer to SI Figs. S11 and S12, which show how mask size and misalignment shift the reconstructed phase vortices.","section":"Supplementary Section A"},{"comment":"The momentum-space step dk and the exact definition of the Gaussian smoothing filter (sigma = 2 pixels in which units) are not specified; without these numbers the numerical derivatives and the reported few-percent error estimate cannot be reproduced.","section":"Methods E, numerical details"},{"comment":"Stating that data and code are available 'upon request' is weaker than current best practice; a permanent repository with versioned code and raw maps would strengthen the reproducibility of the central claims.","section":"Data and Code Availability"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is strong and the primary defect is localized to the quantum-metric extraction formulas; I recommend a major revision rather than rejection. The authors should recompute Fig. 5f and Extended Data Fig. 8 with a gauge-invariant metric, add the missing Hamiltonian-reconstruction formula, and clarify which tight-binding parameters are fitted. I do not see grounds for rejection, and I would be willing to review a revised version."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First, the instrument works: the four intensity maps and the algebraic inversion in Eq. (3) give clean Stokes parameters, and the reconstructed θ, φ windings, eigenstate orthogonality, Hamiltonian coefficients, and sub-linewidth precision near Dirac points are all convincing. This is a real step forward; extending Mueller-matrix polarimetry to the sublattice pseudospin of a continuous photonic lattice with site-selective SLM masks is new, and the experimental care (mask radius, alignment controls in SI) is commendable.\n\nSecond, the quantum metric extraction in Methods E is quantitatively wrong as written. The formulas g_xx = (1 - ⟨u1|u2⟩)/dk_x^2 etc. are complex, gauge-dependent, and actually evaluate to (g_xx + a_x^2)/2, where a_x is the Berry connection, not the quantum metric. So the maps in Fig. 5f and Extended Data Fig. 8 are not measurements of the quantum metric unless the tight-binding simulations used the same erroneous formula (which is not stated). The Berry curvature part, using the Fukui-Hatsugai argument of loop products, is fine. But the real part of the QGT needs a correct, gauge-invariant overlap formula (e.g., taking the real part after fixing the phase of the overlaps, or using the projector formula). This is fixable in a revision, and likely the corrected metric will look similar away from the Dirac points, but as submitted the QGT claim is unsupported.\n\nThe other limitation is that the tight-binding parameters used for the comparison simulations are extracted from the same sample's measured dispersion. That's not a circularity in the Stokes inversion itself, which is a direct linear combination of intensities. But it means the 'agreement with simulations' is partly self-consistency. It would be stronger to present the raw Stokes parameters and a parameter-free comparison, or at least deposit the data and code for independent checking, as the current 'available upon request' is weak.\n\nThe point-like emitter assumption (|k|R_m << 1) is a real approximation, but the SI simulations with different mask sizes and misalignment show that the reconstruction is robust for the chosen R_m = 0.71 µm. I don't count that as a flaw.\n\nWho this is for: people working on photonic band-structure tomography, topological polaritons, and synthetic dimensions. The technique is worth engaging with seriously. My recommendation: send to peer review, but require the authors to fix the quantum metric extraction, re-plot the QGT results, and ideally release data and code.","headline":"A genuinely useful Stokes polarimeter for sublattice pseudospin, with a load-bearing but fixable error in the quantum metric extraction.","tokens_in":21119,"tokens_out":4044,"would_cite":true,"duration_ms":39562,"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":"Four photoluminescence images reconstruct the full Bloch wavefunction of a photonic lattice.","keywords":["sublattice Stokes polarimetry","Bloch eigenstate tomography","quantum geometric tensor","Berry curvature","photonic honeycomb lattice","exciton-polaritons","spatial light modulator","sub-linewidth spectroscopy"],"falsifier":"Measure the reconstructed azimuthal phase-winding positions around the $K$ and $K'$ points while changing the SLM mask radius from $0.71\\,\\mu\\mathrm{m}$ to, say, $1.14\\,\\mu\\mathrm{m}$, or while misaligning the mask by 15--30% of the inter-pillar spacing; if the vortices move away from the Dirac points or the Brillouin zone distorts, the slow-Bloch-phase mapping to a discrete tight-binding model is violated.","tokens_in":19962,"feed_emoji":"🌀","tokens_out":7350,"duration_ms":73373,"temperature":0.7,"pith_summary":"This paper establishes that the sublattice degree of freedom of a bipartite photonic lattice can be measured with a Stokes polarimeter built for pseudo-spin rather than light polarization. Four momentum-resolved photoluminescence images, taken while the two sublattices are left untouched, phase-shifted by π/2, or individually filtered out, are enough to recover the full sublattice Stokes vector of every Bloch mode across the Brillouin zone. From those Stokes vectors the authors reconstruct the Bloch eigenstates in amplitude and phase, obtain the Bloch Hamiltonian, and compute the quantum geometric tensor, Berry curvature, and valley Chern numbers. The same protocol works at the band-touching points of a gapless honeycomb lattice, where a density-matrix analysis separates the overlapping spectral lines and yields eigenenergies with sub-linewidth precision. The significance, if the claim holds, is that full eigenstate geometry of photonic lattices becomes accessible with relatively simple far-field intensity measurements.","feed_headline":"Four intensity images map a lattice's full quantum geometry","feed_subtitle":"A sublattice Stokes polarimeter recovers Bloch modes in amplitude and phase, even at Dirac points.","key_machinery":"The central object is the sublattice Stokes vector, $\\mathbf{S} = (S_0,S_1,S_2,S_3) = (1, \\sin\\theta\\cos\\phi, \\sin\\theta\\sin\\phi, \\cos\\theta)$, which places each Bloch eigenstate on a Poincaré sphere for the two-sublattice spinor. The carrying mechanism is a set of four SLM-implemented Mueller matrices acting on this pseudo-spin: the identity ($A{+}B$), a $\\pi/2$ phase shift between sublattices ($A{+}iB$), and projections onto pure $A$ or pure $B$ emission. The four measured intensities are combined as $I_0 = I_A + I_B$, $I_1 = I_{A+B} - I_A - I_B$, $I_2 = -2I_{A+iB} + I_A + I_B$, and $I_3 = I_A - I_B$, which isolate the Stokes components of every mode. For spectrally overlapping bands, the same observables define a density matrix whose joint diagonalization yields the individual Bloch eigenvectors and their spectral weights.","core_discovery":"The central claim is that sublattice pseudo-spin is as measurable as optical polarization. The paper treats each Bloch state as a spinor $(u_A, u_B)$ mapped to a Stokes vector on a Poincaré sphere whose poles are pure A or pure B sublattice localization. A spatial light modulator placed between crossed polarizers applies controlled phase shifts and attenuations independently to the two sublattices, implementing Mueller matrices for the sublattice degree of freedom: the identity, a quarter-waveplate analogue, and polarizers aligned along the $S_3$ axis. Four measured intensity maps combine linearly to isolate all four Stokes components, giving the polar and azimuthal angles $\\theta$ and $\\phi$ of every eigenstate. The authors demonstrate the method on staggered honeycomb microcavity arrays and on a gapless one, reconstructing the Hamiltonian coefficients $\\lambda_i(k)$, the Berry curvature concentrated at $K$ and $K'$, the quantum metric, and valley Chern numbers, with eigenenergies resolved below the linewidth near the Dirac cones.","pith_inferences":["Applied to the combined polarization and sublattice spinors of exciton-polariton lattices, the same polarimeter would measure a four-component Stokes-like vector, giving direct experimental access to Chern insulating phases.","The paper's own simulations show that mask radius and misalignment move the reconstructed phase vortices, which suggests a built-in calibration check: repeat the $K$-point reconstruction at several mask radii and require the winding positions to stay fixed.","Since the density matrix built from the four intensities encodes non-orthogonality, the method could be pointed at engineered non-Hermitian lattices to track exceptional points instead of assuming orthogonal eigenmodes.","The finite-difference computation of the quantum metric inherits pixel-resolution noise; finer momentum sampling or phase-stepping the SLM masks could push the few-percent uncertainty below the current smoothing limit."],"forward_implications":["The reconstruction yields every coefficient of the Bloch Hamiltonian, including $\\lambda_0(k)$, so hidden couplings such as next-nearest-neighbor tunneling $t'$ become directly measurable rather than inferred from band asymmetry.","With eigenstates known across the Brillouin zone, the full quantum geometric tensor is accessible, giving valley-resolved Berry curvature and Chern numbers that match tight-binding predictions to about 0.01.","At band touching points, joint diagonalization of the measured density matrix separates the overlapping signals of the two bands, resolving linear Dirac dispersions with sub-linewidth energy precision.","Because the measurement protocol is independent of the specific lattice, it can be extended to systems with more internal degrees of freedom, opening multiband topological observables."],"supporting_citations":[{"why":"Supplies the Mueller-matrix formalism of Stokes polarimetry that the sublattice polarimeter generalizes to pseudo-spin.","marker":"[42]"},{"why":"Defines the quantum geometric tensor, the central object reconstructed from the measured eigenstates.","marker":"[6]"},{"why":"Demonstrates Bloch-state tomography via Wilson lines in atomic lattices, the prior art this sublattice polarimeter extends to photonic lattices.","marker":"[17]"},{"why":"Proposes measuring the quantum geometric tensor in two-dimensional photonic and exciton-polariton systems, the goal the experiment realizes.","marker":"[28]"},{"why":"Reports a measurement of the quantum geometric tensor and anomalous Hall drift in exciton-polaritons, showing the class of observables targeted.","marker":"[29]"},{"why":"Explains the physical origin of the next-nearest-neighbor coupling $t'$ through coupling to higher orbitals, used to interpret the measured $\\lambda_0(k)$.","marker":"[45]"}],"fun_headline_variants":["Sublattice Stokes polarimeter: full quantum geometry from four shots","Four intensity images map sublattice spin and reveal Berry curvature","Stokes polarimetry for sublattice polarization in photonic lattices","Measuring the hidden spin of photonic lattices with a Stokes polarimeter","Sublattice pseudo-spin made measurable: a Stokes polarimeter for lattices"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The assumption that the Bloch phase factor $e^{i\\mathbf{k}\\cdot\\mathbf{r}}$ stays nearly constant across each SLM mask region ($|\\mathbf{k}| R_m \\ll 1$, with $R_m = 0.71\\,\\mu\\mathrm{m}$) is load-bearing: if it fails, the filtered emission no longer represents the ideal point-like sublattice spinor and the reconstructed eigenstates are systematically distorted.","fun_headline_variants_meta":{"raw":{"variants":["Sublattice Stokes polarimeter: full quantum geometry from four shots","Four intensity images map sublattice spin and reveal Berry curvature","Stokes polarimetry for sublattice polarization in photonic lattices","Measuring the hidden spin of photonic lattices with a Stokes polarimeter","Sublattice pseudo-spin made measurable: a Stokes polarimeter for lattices"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00132,"raw_usage":{"total_tokens":5405,"prompt_tokens":1003,"completion_tokens":4402,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":619,"completion_tokens_details":{"reasoning_tokens":4309}},"tokens_in":619,"tokens_out":4402,"duration_ms":32102,"temperature":1.0,"reasoning_tokens":4309,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T15:09:32.551157+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the reconstructed azimuthal phase-winding positions around the $K$ and $K'$ points while changing the SLM mask radius from $0.71\\,\\mu\\mathrm{m}$ to, say, $1.14\\,\\mu\\mathrm{m}$, or while misaligning the mask by 15--30% of the inter-pillar spacing; if the vortices move away from the Dirac points or the Brillouin zone distorts, the slow-Bloch-phase mapping to a discrete tight-binding model is violated.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the quantum geometric tensor, the central object reconstructed from the measured eigenstates."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Demonstrates Bloch-state tomography via Wilson lines in atomic lattices, the prior art this sublattice polarimeter extends to photonic lattices."},{"cited_title":"Gianfrate, O","cited_arxiv_id":null,"evidence_quote":"Reports a measurement of the quantum geometric tensor and anomalous Hall drift in exciton-polaritons, showing the class of observables targeted."},{"cited_title":"Mangussi, M","cited_arxiv_id":null,"evidence_quote":"Explains the physical origin of the next-nearest-neighbor coupling $t'$ through coupling to higher orbitals, used to interpret the measured $\\lambda_0(k)$."}],"review_version":1}