{"id":"0324eb9b-f9ae-420a-90eb-0fca3c0288a8","arxiv_id":"2505.11353","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"Trapped polariton condensates show a Hong-Ou-Mandel dip whose shape follows condensate coherence under circular excitation, revives at the Larmor precession frequency under elliptical excitation, and combines with bunching under linear excitation.","lead":"Researchers measured Hong-Ou-Mandel interference between photons emitted by trapped polariton condensates and traced how the interference dip depends on polarization and excitation power. The work adds a new way to probe coherence and spin dynamics in polariton condensates, potential building blocks for future optical simulators and light-matter devices.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Phase-averaging of the non-stabilized interferometer (SI Eq. S5) is the pivotal assumption; all three statistical-regime formulas inherit it, and a direct output-rate balance check can falsify it.","rationale":"The reader's weakest-assumption analysis identified the phase-averaging in SI Eq. S5 and the factorization of higher-order correlations as the critical load-bearing step. My independent review of the derivation confirms that the master equation S7 and its specializations S16, S21, and S27 all depend on dropping the exp(-iω0Δt) terms, and that the normalization is defined as the phase-averaged long-delay product. The mathematical reduction is internally consistent, and the algebraic checks of the circular and linear cases reproduce the published formulas. The elliptical case also survives the algebraic check, and the ad hoc K(τ,Δt) term explicitly does not affect the τ=0 revival, so it is not the central vulnerability. The most serious risk is experimental: whether the assumed uniform phase averaging actually holds in the measured data. The paper does not report output count-rate balance or any independent phase-sampling diagnostic. If the phase is not uniformly sampled, the inferred HOM dip depth and the revival map are systematically biased. This is a concrete, checkable condition, and it is essential to the central claim. Because the reader's verdict of CONDITIONAL already reflects this uncertainty, I recommend no change to the verdict. The suggested test directly targets the assumption and would either validate or falsify the quantitative model.","tokens_in":16373,"tokens_out":42767,"duration_ms":387606,"concrete_test":"Record the time-averaged count rates N1(Δt) and N2(Δt) at the two HOM outputs over the full range of optical delays used in Figs. 1 and 2. Under the phase-averaging assumption, the phase-dependent terms in SI Eq. S4 average to zero, which requires <cos(ω0Δt)> = 0 and hence N1(Δt) = N2(Δt) at every Δt, after correcting for detector efficiencies and dark counts. If the two rates deviate by more than the statistical uncertainty, a non-uniform phase distribution is present; then retain the dropped terms with the measured phase distribution and recompute Eqs. S16, S21, and S27 to see whether the fitted HOM dip and revival shapes change substantially.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—that HOM visibility directly reads out condensate coherence and spinor precession phase matching—rests on reducing the 16-term expansion of the HOM correlation (SI Eq. S4) to the six phase-independent terms of SI Eq. S5 by dropping all terms proportional to exp(-iω0Δt). This drop is valid only if the interferometer optical phase Φ=ω0Δt is uniformly averaged over 2π across the accumulated condensate realizations. The paper asserts that Δt fluctuates in the sub-picosecond range and that millions of realizations ergodically sample the phase, but it provides no direct verification that <cosΦ>≈0 and <sinΦ>≈0. If mechanical drift or slow vibration leaves residual phase coherence, the dropped terms contribute to both the numerator and the long-delay baseline used for normalization, shifting the extracted HOM dip and the revival map in a Δt-dependent way. The same assumption underlies the elliptical and linear cases (Eqs. S21 and S27), so a failure here would invalidate all three statistical-regime claims, not just one. The factorization steps—Isserlis for Gaussian statistics and the common-phase-noise factorization in the spinor model—are additional but secondary assumptions; they are testable separately and are partially validated by the reported agreement. The phase-averaging condition, however, is the single step on which the quantitative meaning of every measured visibility depends.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports Hong-Ou-Mandel (HOM) interferometry measurements on optically trapped polariton condensates under circularly, elliptically, and linearly polarized excitation. The authors derive a general expression for the HOM second-order correlation function in a non-phase-stabilized Mach-Zehnder interferometer (SI Sec. 2, Eqs. S4-S8) and specialize it to three statistical regimes: coherent/Poissonian emission (circular excitation, Eq. S16), spinor Larmor precession with an additional low-frequency modulation (elliptical excitation, Eq. S21), and bunched/thermal emission (linear excitation, Eq. S27). They report that the HOM dip follows the condensate coherence time in the circular case, that the HOM dip revives at the spinor Larmor frequency in the elliptical case, and that the linear case combines polariton bunching with two-polariton interference. The central claim is that HOM visibility provides a direct readout of condensate coherence and spinor-precession phase matching in all three regimes.","tokens_in":16728,"tokens_out":8867,"duration_ms":88244,"significance":"If the results hold, the paper offers a useful extension of HOM interferometry to driven-dissipative polariton condensates and a potentially convenient diagnostic of condensate coherence and spinor dynamics. The theoretical framework is physically motivated, and the coherent-regime derivation (Eq. S16) from classical phase-noise fields is sound; the Gaussian factorization in Eq. S23 is standard, and the linear-regime analysis is partially independent because g(1) is taken from HBT data. The main weaknesses are that the pivotal phase-averaging step in Eq. S5 is unverified, and the elliptical-regime model relies on several fitted or explicitly ad hoc parameters (r, Δω, φ0, Ω, and the K factor), so the quantitative support for the revival claim is weaker than the presentation suggests.","major_comments":[{"comment":"The reduction from the 16-term expansion (Eq. S4) to the six-term expression (Eq. S5) assumes that the interferometer phase Φ=ω0Δt is uniformly sampled over 2π across condensate realizations, so that all terms proportional to e^{±iω0Δt} vanish. This assumption is load-bearing because Eqs. S16, S21, and S27, as well as the normalization G_HOM(∞,Δt), all inherit Eq. S5. The manuscript states that Δt fluctuates in the sub-picosecond range and that the system is ergodic, but it provides no direct experimental verification that ⟨e^{iΦ}⟩≈0 and ⟨e^{-iΦ}⟩≈0. If slow mechanical drift leaves residual phase coherence, the dropped terms contribute to both the numerator and the long-delay baseline, shifting the extracted dip depth and revival map in a Δt-dependent way. Please provide a direct check—for example, the measured long-τ baseline as a function of Δt, or a comparison of the averaged data with the stabilized-phase expression Eq. S28—or specify the interferometer stability and demonstrate that it guarantees uniform phase sampling.","section":"SI Sec. 2, Eq. S5"},{"comment":"The elliptical-regime model is not yet a predictive test of the claimed HOM revival. The Larmor frequency Δω, the mode imbalance r, and the slow-modulation parameters φ0 and Ω are free parameters adjusted to the HOM color map, and the text explicitly states that the K factor is introduced 'to better align the experimental data with the theory.' In addition, although the text says the width of Δω is responsible for dephasing of the spin precession, no damping factor for that dephasing appears in Eq. S21. The revival claim would be substantially strengthened if Δω were fixed from the independently measured HBT Larmor oscillations (Fig. S4) and the HOM revival map then predicted, with a specified dephasing envelope, rather than fitted together with the other parameters.","section":"SI Sec. 2B, Eq. S21"},{"comment":"The central quantitative claim that the HOM dip follows the condensate coherence time rests on fits of τ0 and γ, but no fitted parameter values, uncertainties, or goodness-of-fit metrics are reported for the color maps in Figs. 1 and 2 or for the trap-size comparison in SI Sec. 4. Please report the fitted τ0 and γ values with confidence intervals and provide a quantitative comparison (e.g., residuals or χ² per degree of freedom) for the main HOM curves. This is needed for the reader to assess how strongly the data constrain the model and whether the elliptical revival periodicity is consistent with the independently measured Larmor frequency within uncertainty.","section":"Main text Eqs. 7/11; SI Secs. 2A/2C"}],"minor_comments":[{"comment":"Eq. S26 writes G_HOM^(2) with a prefactor |E0|^2, but as a fourth-order correlation function it should be |E0|^4; this dimensional typo should be corrected.","section":"SI Eq. S26"},{"comment":"The citation numbering is inconsistent: references [3], [4], and [5] are used for multiple distinct works, and some entries appear in non-sequential order; the list should be renumbered.","section":"Reference list"},{"comment":"The color maps in Figs. 1 and 2 lack color bars and the abbreviation 'N.C.' is undefined in the captions; please add a color scale and define the normalization.","section":"Figures 1 and 2"},{"comment":"The step from the qualitative polarization-correlation expression in Eq. S32 to the final HOM visibility formula in Eq. S33 is not shown; please add a brief derivation or an explicit statement of the approximations used.","section":"SI Sec. 3, Eq. S33"}],"recommendation":"major_revision","confidential_remarks":"The phase-averaging assumption in SI Eq. S5 is the central risk. If the authors can supply a direct experimental check of the baseline and fix the elliptical model's parameters using independent HBT data, the paper could become acceptable. The duplicate references and the ad hoc K factor should also be addressed before resubmission."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nShort version: this is a worthwhile experimental paper, not a breakthrough. It applies HOM interferometry to optically trapped polariton condensates for the first time, as far as the citation record shows, and reports one genuinely new thing: under elliptical pumping, the HOM dip revives at the spinor Larmor precession frequency. The theoretical model is assembled from standard classical coherence theory—no new formalism—but it is applied carefully and the SI derivation is clear.\n\nWhat the paper does well: the circular-polarization case shows the HOM dip tracking the coherence time and approaching the classical 1/2 limit, consistent with a stable single-mode condensate. The linear-polarization case is the most convincing: they measure g(1) from HBT data, plug it into Eq. 11, and the HOM curve comes out essentially right without extra fitted parameters. That is a real success and a good example of how HOM can be used as a characterization tool. The Larmor revival observation is visually striking and the connection to precession phase matching is physically plausible. The SI is detailed and mostly honest about what is modeled and what is measured.\n\nSoft spots, in proportion: the stress-test concern about SI Eq. S5 is legitimate. All three formulas drop terms proportional to exp(-i ω0 Δt) on the claim that the non-stabilized interferometer's delay fluctuates enough to average the phase. That is a standard assumption in classical HOM, and the sub-picosecond fluctuation claim is plausible if it means the phase changes by many radians, but they never verify <cosΦ>≈0 directly. A failure there would shift the dip and the normalization in every regime. I don't think it fails—the linear-case agreement would be hard to reproduce otherwise—but a direct check, e.g., detector count-rate balance vs. Δt, would close it. More concrete: the circular and elliptical cases involve fitted parameters, and the elliptical revival is built on several of them (τ0, r, Δω, φ0, Ω), with the K factor explicitly ad hoc to improve agreement. The revival periodicity is already present in the HBT data, so the HOM revival map is partly a restatement of the Larmor frequency plus a modulation term. That makes the linear case the solid core; the elliptical case is suggestive but not quantitatively pinned down.\n\nBottom line: this deserves a serious referee. It is a legitimate experimental advance within the polariton field, with one clean independent test and one new observation. It would be stronger with raw data, error bars, and a test of the phase-averaging assumption, but those are revision-level issues, not desk-reject issues. I'd send it to review and ask for those additions.\n\nYours,","headline":"A solid experimental HOM study on trapped polariton condensates with a genuinely new revival observation; the model is classical and the linear case is the most convincing, but the elliptical revival rests on fitted parameters and the phase-averaging assumption in the SI is not directly tested.","tokens_in":17237,"tokens_out":4058,"would_cite":true,"duration_ms":40929,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"In trapped polariton condensates, the Hong-Ou-Mandel dip follows the coherence time and revives at the spinor Larmor frequency.","keywords":["Hong-Ou-Mandel interference","polariton condensate","second-order coherence","Larmor precession","spinor condensate","photon statistics","optical trap","two-photon interference"],"falsifier":"Stabilize the interferometer while keeping the optical delay fixed and scan the phase $\\Phi$: the model predicts the HOM dip should oscillate with $\\cos(2\\Phi)$ according to Eq. S28, whereas the non-stabilized formulas contain no such phase dependence. If the dip shows no phase dependence, or if an independently measured $g^{(1)}(\\tau)$ fails to reproduce the measured dip depth through Eqs. S16, S21, or S27, the phase-averaging or factorization assumption is wrong.","tokens_in":16162,"feed_emoji":"🔬","tokens_out":8978,"duration_ms":88121,"temperature":0.7,"pith_summary":"The paper reports Hong-Ou-Mandel (HOM) two-photon interference measurements on an optically trapped polariton condensate and shows that the HOM dip—the drop in coincidence counts when two indistinguishable photons meet at a beam splitter—tracks the condensate's first-order coherence time. For a circularly polarized condensate, the dip approaches the classical limit set by the squared first-order coherence, confirming coherent, spin-stable emission. Under elliptical excitation, spinor Larmor precession modulates the interference, and the HOM dip revives periodically when the interferometer's optical delay matches integer precession periods. For linear excitation, the suppressed polarization mode emits bunched, thermal-like light, and the measured correlation combines photon bunching with two-photon interference, giving a dip twice as deep as in the coherent case. A general phase-averaged formula for the non-stabilized interferometer unifies the three regimes, making HOM visibility a direct readout of condensate coherence and spinor dynamics.","feed_headline":"Two-photon interference tracks polariton condensate coherence","feed_subtitle":"In trapped polariton condensates, the HOM dip follows coherence time and revives at the spinor Larmor frequency.","key_machinery":"The central object is the second-order correlation function $G^{(2)}_{\\mathrm{HOM}}(\\tau,\\Delta t)$ at the two outputs of an unbalanced, non-phase-stabilized interferometer; after averaging over rapid optical-delay fluctuations, the sixteen-term expansion reduces to a compact expression involving only the intensity autocorrelation $G^{(2)}(\\tau)$ and one four-time correlation term. The emission field is treated as stationary and ergodic, and in each statistical regime the four-time term is evaluated through $g^{(1)}$: exactly for Poissonian coherent fields, through the Isserlis/Gaussian factorization for thermal fields, and through a two-frequency spinor ansatz with a Larmor beat for the precessing condensate. The quantity $t_{\\mathrm{min}} = \\min(|\\Delta t|, |\\tau|)$ carries the coherence-time dependence of the dip, while the revival condition is set by the phase factor $\\cos(\\Delta\\omega \\Delta t/2)$ and the periodicity of the spinor precession.","core_discovery":"The central discovery is that the two-photon HOM correlation function of a trapped polariton condensate is governed by the same first-order coherence function $g^{(1)}$ in all three excitation regimes, once the interferometer phase is averaged. For circularly polarized coherent emission, the normalized coincidence rate is $g^{(2)}_{\\mathrm{HOM}}(\\tau,\\Delta t) = \\left(1 - \\frac{1}{2}|g^{(1)}(t_{\\mathrm{min}})|^2\\right)/\\left(1 - \\frac{1}{2}|g^{(1)}(\\Delta t)|^2\\right)$, so the dip at zero electrical delay deepens with condensate coherence and saturates at the classical value of $1/2$ when the optical delay far exceeds the coherence time. For elliptically polarized excitation, the two circular spin components beat at the Larmor frequency $\\Delta\\omega$, adding a $\\cos(\\Delta\\omega\\tau)$ term and producing a revival of the HOM dip whenever $\\Delta t$ is an integer multiple of the precession period. For linearly polarized excitation, the filtered mode has Gaussian thermal statistics, and the HOM signal combines photon bunching with two-polariton interference, with the absolute dip twice as pronounced as in the Poissonian case. The model reproduces the measured correlation maps, including the absence of anti-correlation at zero optical delay and the revival periodicity.","pith_inferences":["Because the optical-delay revival period is set by the Larmor frequency, the same setup could serve as a self-referenced precession clock: the revival spacing directly yields $\\Delta\\omega$ without an external time base.","The additional low-frequency modulation $K(\\tau,\\Delta t)$ introduced at high pump powers hints that stabilized HOM correlation maps could also be sensitive to mechanical or time-crystal-like oscillations of the microcavity, though this goes beyond what the paper proves.","A tunable-ellipticity scan across the revival conditions would test whether the dip envelope stays fixed by $g^{(1)}(\\Delta t)$ or acquires extra spin-dephasing structure; the current model places all spin dephasing in the width of $\\Delta\\omega$, which is a testable distinction."],"forward_implications":["For a circularly polarized condensate, the HOM dip at zero electrical delay is determined entirely by $|g^{(1)}(t_{\\mathrm{min}})|^2$, so scanning the optical delay maps the condensate coherence time without a separate spectral linewidth measurement.","When the condensate spinor precesses, HOM visibility revives only near optical delays that are integer multiples of the Larmor period, making the revival spacing a direct readout of the precession frequency.","For the linearly polarized bunched mode, the HOM correlation is a superposition of thermal bunching and two-polariton interference, and the absolute depth of the dip is twice that of the Poissonian case.","Stabilizing the interferometer at a fixed phase $\\Phi$ should change the averaged non-stabilized result into an expression whose dip oscillates with $\\cos(2\\Phi)$, giving an experimental control knob for two-photon interference."],"supporting_citations":[{"why":"Defines the Hong-Ou-Mandel two-photon interference effect that the measurements implement.","marker":"[1]"},{"why":"Earlier realization of the three photon-statistics regimes in a spinor polariton condensate that this experiment extends to HOM interferometry.","marker":"[4]"},{"why":"Establishes non-equilibrium Bose-Einstein condensation of exciton-polaritons and the coherence properties assumed for the condensate.","marker":"[20]"},{"why":"Characterizes spin stability and coherence of optically trapped polariton condensates, the baseline against which the circular-polarization result is compared.","marker":"[22]"},{"why":"Supplies the effective out-of-plane magnetic field and Larmor precession mechanism for the spinor condensate that produces the HOM revivals.","marker":"[26]"},{"why":"Gives the classical-source limit of the HOM dip as $1 - \\frac{1}{2}|g^{(1)}|^2$, which the circular case reproduces.","marker":"[29]"},{"why":"Provides the quasi-classical phase-noise description and Gaussian correlator factorization used to derive the regime-specific formulas.","marker":"SI [2]"},{"why":"Supplies the Isserlis-type factorization of higher-order correlations for Gaussian statistics, used for the linearly polarized thermal case.","marker":"SI [5]"}],"fun_headline_variants":["Polariton HOM dip revives at Larmor frequency","Coherence sets two-photon interference in polariton condensates","Spinor revival of Hong-Ou-Mandel dip in trapped condensates","HOM interferometry reveals polariton coherence and spinor beats","Two-polariton interference follows condensate coherence time"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the interferometer's optical path difference fluctuates rapidly enough to wipe out all optical-phase terms, while the condensate emission is a stationary ergodic classical field whose higher-order correlations factor through $g^{(1)}$ (Poissonian or Gaussian).","fun_headline_variants_meta":{"raw":{"variants":["Polariton HOM dip revives at Larmor frequency","Coherence sets two-photon interference in polariton condensates","Spinor revival of Hong-Ou-Mandel dip in trapped condensates","HOM interferometry reveals polariton coherence and spinor beats","Two-polariton interference follows condensate coherence time"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000189,"raw_usage":{"total_tokens":1332,"prompt_tokens":940,"completion_tokens":392,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":556,"completion_tokens_details":{"reasoning_tokens":306}},"tokens_in":556,"tokens_out":392,"duration_ms":3664,"temperature":1.0,"reasoning_tokens":306,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T20:53:57.370747+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Stabilize the interferometer while keeping the optical delay fixed and scan the phase $\\Phi$: the model predicts the HOM dip should oscillate with $\\cos(2\\Phi)$ according to Eq. S28, whereas the non-stabilized formulas contain no such phase dependence. If the dip shows no phase dependence, or if an independently measured $g^{(1)}(\\tau)$ fails to reproduce the measured dip depth through Eqs. S16, S21, or S27, the phase-averaging or factorization assumption is wrong.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Earlier realization of the three photon-statistics regimes in a spinor polariton condensate that this experiment extends to HOM interferometry."},{"cited_title":"Baryshev, A","cited_arxiv_id":null,"evidence_quote":"Establishes non-equilibrium Bose-Einstein condensation of exciton-polaritons and the coherence properties assumed for the condensate."},{"cited_title":"Kasprzak, R","cited_arxiv_id":null,"evidence_quote":"Supplies the effective out-of-plane magnetic field and Larmor precession mechanism for the spinor condensate that produces the HOM revivals."},{"cited_title":"Cilibrizzi, A","cited_arxiv_id":null,"evidence_quote":"Gives the classical-source limit of the HOM dip as $1 - \\frac{1}{2}|g^{(1)}|^2$, which the circular case reproduces."}],"review_version":1}