{"id":"73688fb0-8de6-4315-a9d0-ea4677bccbb5","arxiv_id":"2608.03658","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"A waveguide-coupled two-level emitter driven by pulsed squeezed vacuum can produce two-mode non-Gaussian states from which high-fidelity squeezed cat states are obtained by temporal-mode selection and single-photon heralding.","lead":"A single two-level emitter in a waveguide, hit by a squeezed light pulse, produces a two-part non-Gaussian light state. Selecting the right temporal modes and detecting one photon in one part can herald a large squeezed cat state in the other part with about 14% probability.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Two-mode truncation plus pure-state approximation breaks the parity herald; a third-mode simulation is needed to validate the 99.3% squeezed-cat fidelity.","rationale":"The reader's weakest assumption—the two-mode truncation and pure-state approximation—is the same load-bearing point I identify. The physics is otherwise plausible: the coherent-drive Wigner negativity is a simple saturation effect, the squeezed-drive bimodal structure is a natural extension, and the chirality-recovery beam-splitter transformation is algebraically clean. The numerical protocol is described with enough detail (QuTiP, N=13, virtual-cavity construction) to be reproducible. However, the headline result is a set of quantitative numbers: 99.3% fidelity to a squeezed odd cat with |α|^2=3.5, stellar rank >=3, and ~15% heralding probability. All of these are computed inside a two-mode Hilbert space after replacing a 99.5%-pure density matrix with a pure state. The missing 0.3% of the output population is not automatically harmless: because the herald is a parity measurement, a small amplitude in modes outside v1,v2 can flip the parity correlation between the two selected modes and contaminate the accepted branch. The paper explicitly acknowledges the approximation ('we will therefore approximate it as a pure state') but does not bound its effect on the parity-resolved quantities. A third-mode simulation is the natural and decisive check: it directly quantifies the leakage into the next eigenmode and its effect on the heralded state. If the check passes, the central claim stands; if it fails, the protocol may still work but the quoted fidelities, ranks, and rates need revision. The reader's conditional verdict already captures this uncertainty, so I do not recommend changing the verdict.","tokens_in":15501,"tokens_out":12410,"duration_ms":156466,"concrete_test":"Augment the virtual-cavity simulation to include a third output cavity capturing the next eigenmode v3 of g(t1,t2), keeping the same N=13 truncation and applying the same θ3 rotation between v1 and v2. Compute (i) the v3 population; (ii) the reduced state of mode v1 after projecting mode v2 onto odd parity and onto |1><1|, tracing out v3; (iii) the fidelity of that state with the target squeezed odd cat S(8 dB)(|α>-|-α>), α^2=3.5, and the Appendix B stellar-rank witness. If fidelity remains >=99.3%, rank stays >=3, and the opposite-parity weight (n1 even, n2 odd) is <0.1%, the pure two-mode approximation is validated. If the fidelity drops by more than ~1% or the rank bound falls, the headline numbers must be corrected.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section IV restricts the output to the two dominant modes v1,v2 (99.7% of the output population) and approximates the joint state as pure (purity 99.5%). The parity decomposition in Eq. (1) then writes |Psi>_{1,2} = a|even>_1|even>_2 + b|odd>_1|odd>_2, which is exact only inside H_v1⊗H_v2. In the full state, total photon-number parity is even, so n_1+n_2 has the same parity as n_rest. The 0.3% of photons outside the two modes can therefore populate the opposite-parity sectors (even,odd) and (odd,even) whenever n_rest is odd. An odd-parity or |1><1| projection on mode v2 then accepts an admixture of the desired odd-odd sector with this opposite-parity sector, which is entirely absent from the pure two-mode calculation. Since the claimed fidelity (99.3%) and purity (0.987) leave only about one percent of slack, while the leakage/impurity are 0.3–0.5%, the central quantitative claims—squeezed-cat fidelity, stellar rank >=3, and the 15% heralding rate—are not yet robust. The parity selection is a discrete, nonlinear postselection, so small leakage can enter at amplitude level rather than as a simple population correction.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies non-Gaussian state generation from a single two-level emitter in a waveguide, driven by pulsed coherent or squeezed vacuum inputs. Using the virtual-cavity input-output formalism, the authors identify two dominant output temporal modes, optimize a linear mode rotation, and characterize the resulting two-mode state by purity, Wigner negativity, core-state structure, and stellar-rank witnesses. For coherent pulses they report deterministic Wigner-negative states with core confined to {|0>,|1>} and stellar rank 1. For squeezed vacuum pulses they report a bimodal non-Gaussian state; in a particular rotated-mode basis, projecting one mode onto an odd-parity or single-photon state heralds in the other mode a squeezed odd cat state with size |alpha|^2 = 3.5, squeezing 8 dB, fidelity ~99.3%, stellar rank >=3, and success probability ~15%. The paper also analyzes losses and proposes a beam-splitter arrangement that restores an effective chiral interaction in a bidirectional waveguide.","tokens_in":15867,"tokens_out":18311,"duration_ms":205310,"significance":"If the quantitative claims survive scrutiny, this would be a useful proposal: a deterministic Gaussian drive plus a single two-level nonlinearity, followed by temporal-mode sorting and PNR heralding, could provide high-rate non-Gaussian resources relevant for cat and GKP state preparation. The chiral-recovery beam splitter is a clean and potentially practical idea. The numerical work is described with explicit Hilbert-space dimensions, and the parity decomposition is physically transparent. However, the central fidelity and rank claims rest on a two-mode pure-state approximation whose error is not controlled; this is the main obstacle to accepting the results as stated.","major_comments":[{"comment":"The parity decomposition assumes the output is confined to v1⊗v2. The statement that n1 and n2 must have the same parity is only true modulo n_rest; since total parity is even, n1+n2 has the parity of n_rest. With only 99.7% of the population in v1,v2 and purity 99.5%, the pure-state approximation discards opposite-parity sectors (even,odd) and (odd,even) that arise when n_rest is odd. A |1><1| projection on mode 2 can then accept these sectors. A Markov bound gives P(n_rest≥1) ≤ ⟨n_rest⟩ ≈ 0.0035; relative to the 14–15% herald probability this is a worst-case conditional contamination of ≈2.5%, larger than the 0.6–0.7% infidelity implied by the claimed fidelities. A third-mode simulation or a rigorous conditional-state error bound is needed.","section":"Section IV, Eq. (1)"},{"comment":"The headline results (99.3% fidelity for the θ3 odd component; stellar rank ≥3) are computed in the purified two-mode model. The purity 99.5% of ρ1,2 does not by itself bound the error of the herald-conditioned state, because postselection can amplify small components. The paper should either include the third mode in the simulation or provide an explicit amplitude-level bound. Until then, the quantitative claims are not established.","section":"Section IV.B / Table I"}],"minor_comments":[{"comment":"The phrase \"We’ll focus here also on one set of parameters\" is informal; also, the approximation \"we will therefore approximate it as a pure state\" should be accompanied by at least a quantitative statement of the fidelity between ρ1,2 and the adopted pure state.","section":"Section IV"},{"comment":"The beam-splitter parameters t and r conflict with time t and stellar rank r*; consider renaming them (e.g., η, ξ) to avoid ambiguity.","section":"Section V.B, Eq. (5)"},{"comment":"\"Scarce matrix\" should be \"sparse matrix\".","section":"Appendix B"},{"comment":"The condition F > F*_k ⇒ r* ≥ k+1 is only stated in Appendix B; a one-line reminder in the caption would make the table self-contained.","section":"Table I"},{"comment":"The loss analysis is performed for coherent input, while the heralded squeezed-vacuum protocol is the main resource; the robustness of the latter to loss is not shown. A sentence explaining the intended transferability would be helpful.","section":"Section V.A"},{"comment":"Use \"non-Gaussian\" consistently; the lowercase \"non-gaussian\" in the abstract should be corrected.","section":"Abstract"}],"recommendation":"major_revision","confidential_remarks":"I agree with the reader's conditional verdict. The two-mode truncation and pure-state approximation are load-bearing for the headline squeezed-cat fidelity and stellar-rank claims. The authors should be asked to add either a three-output-mode simulation (even with a reduced photon-number cutoff) or a rigorous bound on the herald-conditioned state error. If the contamination is shown to be below ~0.1% in the conditional state, the paper could be acceptable; otherwise the claims need to be scaled back. The chiral-recovery section is an original and useful addition."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a genuine new regime, not a repackaging. The authors drive a single TLS in a waveguide with pulsed squeezed vacuum, identify a structured two-mode output, and show that a mode rotation plus parity or single-photon heralding on one mode can yield a large squeezed odd cat state in the other, with claimed fidelity around 99%, stellar rank at least 3, and a heralding probability around 15%. The physics is plausible: a TLS saturates at one excitation, and the squeezed input gives the bimodal structure they exploit. The formal machinery I checked is standard and clean: input-output plus virtual-cavity formalism, explicit Hilbert-space dimensions (d=4394), and the interferometric scheme to recover an effective chiral interaction from a bidirectional waveguide is a nice standalone result. The paper is also honest that the core-state search is not proven optimal, which is a real but secondary caveat.\n\nThe main soft spot is exactly where the stress-test note lands. In Section IV the authors restrict to the two dominant modes (99.7% of output population) and purify a 99.5%-pure joint state, then use the parity decomposition as if it were exact. In the full state, total photon-number parity is even, so n1+n2 has the same parity as n_rest. The 0.3% of photons outside the two modes can therefore populate the (even,odd) and (odd,even) sectors whenever n_rest is odd. A single-photon or odd-parity projection on mode 2 accepts those sectors, and that admixture is absent from the truncated calculation. Since the claimed fidelity (99.3%) and purity (0.987) leave only about a percent of slack, the quantitative claims—including the stellar-rank bound and the 14–15% rate—are not yet robust. This does not kill the qualitative proposal, but it means the headline numbers are unvalidated until the authors do a third-mode simulation or otherwise bound the parity-leakage terms. The heuristic core-state witness is a smaller soft spot: the rank bounds inherit the uncertainty of an empirical loss minimization, though the witnesses may still be conservative.\n\nThe citation pattern looks fine; the paper engages with the relevant recent waveguide-QED and cat-generation work. The estimated rates in the 100 kHz–MHz range are ideal-model estimates, conditional on QPG efficiency and PNR detection, and the authors say so.\n\nWho is this for? People working on waveguide-QED sources of non-Gaussian light and on cat/GKP resource generation. It deserves a serious referee. A good report should ask for convergence checks and a loss budget, not a rejection. I would bring it to reading group.","headline":"A genuinely new single-emitter route to squeezed cat states, but the headline numbers rest on an unvalidated two-mode truncation and need a third-mode check before I would trust them.","tokens_in":16332,"tokens_out":3035,"would_cite":true,"duration_ms":37274,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["42.50.Ex","03.67.-a"],"model":"deepseek-v4-flash","headline":"A single two-level emitter driven by squeezed vacuum can herald large squeezed cat states, with |α|²=3.5 and stellar rank ≥3, at 15% success probability.","keywords":["non-Gaussian light","waveguide quantum electrodynamics","squeezed vacuum","Schrödinger cat states","stellar rank","Wigner negativity","temporal modes","quantum pulse gate"],"falsifier":"Set up a waveguide-coupled two-level emitter driven by an 8 dB squeezed-vacuum pulse with γτ = 0.25, separate the two dominant output temporal modes with a quantum pulse gate, and perform two-mode homodyne tomography conditioned on single-photon detection in one mode. The central claim would be contradicted if the two modes account for substantially less than 99.7% of the output population, if the joint state's purity is far below 99.5%, or if the heralded mode's Wigner function does not show the three negative lobes and ~99% fidelity to a squeezed odd cat with |α|² ≈ 3.5.","tokens_in":15413,"feed_emoji":"🐈","tokens_out":10553,"duration_ms":115376,"temperature":0.7,"pith_summary":"This paper tries to show that a single two-level emitter in a waveguide—the simplest possible quantum nonlinearity—can turn Gaussian light into genuinely non-Gaussian states useful for optical quantum computing. For a pulsed squeezed-vacuum input, the output field splits into two dominant temporal modes whose photon-number parities are locked together, and this two-mode state can be harvested: detecting a single photon in one mode leaves the other mode in a large squeezed Schrödinger-cat state, with stellar rank at least 3 and near-99% fidelity to an ideal odd cat of size |α|² = 3.5, at about 15% success probability. The same emitter driven by a coherent pulse already makes deterministic Wigner-negative states, but only rank-1 states of the vacuum/single-photon type. The paper also shows Wigner negativity survives down to 50% coupling efficiency and that a simple interferometer can restore an effectively chiral interaction in a bidirectional waveguide. If right, this offers a comparatively simple, high-repetition-rate route to non-Gaussian resources such as cat states for GKP error correction.","feed_headline":"Squeezed light plus one emitter yields 99% cat states","feed_subtitle":"A photon click on one temporal mode heralds a rank-3 squeezed odd cat at 15% probability, easing GKP-state production.","key_machinery":"The load-bearing object is the joint density matrix of the two dominant output temporal modes, decomposed into even- and odd-parity branches. Squeezed vacuum populates only even Fock states and the emitter interaction conserves total photon number, so the parity of the two output modes is locked together; this is what lets a parity or single-photon measurement on one mode herald a specific non-Gaussian branch in the other. The authors rotate the mode basis with a virtual beam splitter and use the virtual-cavity formalism to compute the states, then certify non-Gaussianity through k-robustness stellar-rank witnesses.","core_discovery":"A single two-level emitter driven by a pulsed squeezed vacuum creates a two-mode output state whose photon-number parities are locked: both modes are even together or odd together. Selecting the output mode basis and detecting one photon in one mode picks the odd branch, which the paper identifies as a squeezed odd Schrödinger cat with |α|² = 3.5, 99.3% fidelity, stellar rank at least 3, and 15% heralding probability; projecting exactly onto one photon gives a similar cat at |α|² = 3.1, 98.7% purity, 99.4% fidelity, and 14% probability. Coherent pulses give deterministic Wigner-negative states, but only within the vacuum/single-photon subspace.","pith_inferences":["The paper leaves open whether the two-mode state itself, rather than a single heralded mode, can be consumed directly as a resource; if its even and odd branches are correlated in the right way, parity measurements on one mode could implement the breeding step of GKP-state preparation without first collapsing to a single mode.","Because the mode rotation is done numerically as a virtual beam splitter, an experimental implementation could tune the temporal-mode basis in real time; a natural extension is to search over input pulse shapes, not just mode bases, to push the odd-cat branch above 15%.","The near-overlap between one optimized output mode and the incoming pulse shape suggests a simplified variant: use the transmitted pulse mode directly and rely on photon-number-resolving detection alone, accepting lower Wigner negativity to avoid the quantum pulse gate."],"forward_implications":["Coherent drive: within the explored parameter range, the deterministic output is a high-purity displaced superposition of |0⟩ and |1⟩, with Wigner negativity volume up to 88% of a single photon's and a core state confined to the two lowest Fock levels.","Squeezed drive, deterministic path: in the mode basis that maximizes single-mode purity, the output is a squeezed even cat state (purity 0.86), whose even-parity component reaches 99.4% fidelity to an ideal squeezed even cat.","Squeezed drive, heralded path: a photon-number parity or single-photon detection on the second temporal mode produces a squeezed odd cat with |α|² = 3.5 at 15% probability and 99.3% fidelity, or with |α|² = 3.1 and purity 0.987 under exact |1⟩⟨1| projection at 14% probability.","Loss and directionality: Wigner negativity of the coherent-pulse output survives until about 50% total coupling efficiency; a beam-splitter arrangement recombining left- and right-moving fields turns a bidirectional waveguide into an effectively chiral one, so the 50% collection limit can be bypassed when the relative phase is stabilized."],"supporting_citations":[{"why":"Supplies the virtual-cavity formalism that converts pulsed input and output modes into discrete cavities for the numerical simulation.","marker":"[25, 26]"},{"why":"Provides the quantum pulse gate that separates and converts the chosen temporal modes into distinct spatial or frequency modes.","marker":"[27]"},{"why":"Analyzes the temporal-mode structure of light scattered by a two-level system, motivating the two-mode decomposition used here.","marker":"[16]"},{"why":"Demonstrates deterministic photon subtraction in waveguide QED, the earlier mechanism closest to the single-photon heralding used here.","marker":"[13]"},{"why":"Defines the k-robustness witnesses used to certify stellar-rank lower bounds for the generated states.","marker":"[46]"},{"why":"Shows squeezed-drive excitation of an emitter can reach higher-Fock core states, the strategy extended to pulsed squeezed vacuum here.","marker":"[19]"},{"why":"Demonstrates cat-to-GKP breeding, the target application that large squeezed cat states would feed.","marker":"[38]"}],"fun_headline_variants":["One emitter plus squeezed light: 99% cat states","Heralded odd cats at 15% odds from a single emitter","Parity-locked photons yield rank-3 Schrödinger cats","Single two-level system makes large squeezed cats","Waveguide single emitter: 99% fidelity cat states"],"cache_read_input_tokens":2688,"weakest_assumption_plain":"The load-bearing approximation, stated in Section IV, is that the two dominant output modes capture 99.7% of the population and that the joint state can be treated as pure at 99.5% purity; if the discarded 0.3% or the 0.5% impurity breaks the parity correlations, the quoted fidelities and stellar ranks shift.","fun_headline_variants_meta":{"raw":{"variants":["One emitter plus squeezed light: 99% cat states","Heralded odd cats at 15% odds from a single emitter","Parity-locked photons yield rank-3 Schrödinger cats","Single two-level system makes large squeezed cats","Waveguide single emitter: 99% fidelity cat states"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000231,"raw_usage":{"total_tokens":1299,"prompt_tokens":699,"completion_tokens":600,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":443,"completion_tokens_details":{"reasoning_tokens":518}},"tokens_in":443,"tokens_out":600,"duration_ms":8327,"temperature":1.0,"reasoning_tokens":518,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T15:13:15.266955+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Set up a waveguide-coupled two-level emitter driven by an 8 dB squeezed-vacuum pulse with γτ = 0.25, separate the two dominant output temporal modes with a quantum pulse gate, and perform two-mode homodyne tomography conditioned on single-photon detection in one mode. The central claim would be contradicted if the two modes account for substantially less than 99.7% of the output population, if the joint state's purity is far below 99.5%, or if the heralded mode's Wigner function does not show the three negative lobes and ~99% fidelity to a squeezed odd cat with |α|² ≈ 3.5.","supporting_citations":[{"cited_title":"Gonz´ alez-Tudela, V","cited_arxiv_id":null,"evidence_quote":"Analyzes the temporal-mode structure of light scattered by a two-level system, motivating the two-mode decomposition used here."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Demonstrates deterministic photon subtraction in waveguide QED, the earlier mechanism closest to the single-photon heralding used here."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the k-robustness witnesses used to certify stellar-rank lower bounds for the generated states."},{"cited_title":"Pasharavesh and M","cited_arxiv_id":null,"evidence_quote":"Shows squeezed-drive excitation of an emitter can reach higher-Fock core states, the strategy extended to pulsed squeezed vacuum here."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Demonstrates cat-to-GKP breeding, the target application that large squeezed cat states would feed."}],"review_version":1}