{"id":"907e9273-7abb-4553-8408-131d26cc01a9","arxiv_id":"2507.04829","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A second-quantized multi-Lambda cavity model yields an effective Raman beam-splitter Hamiltonian that predicts Hong-Ou-Mandel bunching for both atoms and photons.","lead":"The authors derive a second-quantized effective Hamiltonian for multi-level atoms in a two-mode cavity, including center-of-mass motion, counter-rotating terms, and an effective atom-atom interaction. They use it to show that Raman diffraction acts as a beam splitter that can produce Hong-Ou-Mandel bunching for atoms and photons.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The formal averaging step that produces H_eff is internally inconsistent: the dissipator D2 in Eq. (B20) is identically zero as written, while Eqs. (B30)-(B32) treat it as nonzero with singular 1/omega(-) coefficients; until this algebra is repaired, Eq.","rationale":"The reader identified the low-pass frequency hierarchy as the weakest assumption and noted the apparent error in Eq. (B20), but did not treat that algebraic inconsistency as the load-bearing issue. I agree with the reader that the timescale separation in Appendix B.2 is a real premise; however, the more fundamental checkpoint is the internal algebra of the averaging method, because the effective Hamiltonian is extracted from that algebra. If D2 as defined in Eq. (B20) is identically zero while Eqs. (B30) and (B32) use it as nonzero, the central prediction has no verified foundation. This is not an accusation of dishonesty; it may be a typographical or sign error, and the final H_eff may survive a corrected derivation. Still, with no code, data, or machine-checked proof, an unverified central derivation is exactly what a conditional verdict should rest on. I also checked the two-particle Hong-Ou-Mandel reasoning under the assumption that Eq. (11) is valid: the non-bunching amplitude scales as cos(2 theta Omega_n), so the stated 50:50 condition indeed leaves only the bunching outputs. The problem is therefore upstream, not in the application. The verdict should remain conditional: acceptance requires the symbolic recomputation to validate Eqs. (B17)-(B32) or a corrected appendix that establishes the same effective Hamiltonian.","tokens_in":44354,"tokens_out":18819,"duration_ms":235899,"concrete_test":"Perform an independent symbolic recomputation of the second-order generator L2 using Eqs. (B8), (B11), and (B16) for the harmonic Hamiltonian Eq. (B21), then compare the result with the decomposition in Eqs. (B17)-(B20). Specifically, check whether D2 as printed is identically zero, and whether the n=m diagonal terms in Eq. (B32) cancel when computed directly from Eq. (B29) without passing through the singular 1/omega(-) notation. If the recomputation reproduces Eq. (11), the concern is resolved; if not, the derivation of H_eff needs repair before the beam-splitter and Hong-Ou-Mandel claims can be accepted.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The load-bearing step is the passage from the second-order time-averaging expansion, Eqs. (B16)-(B19), to the effective-Hamiltonian form used in Eq. (11). As printed, Eq. (B20) defines D2[O] ≡ H O U1 - H O U1 + U1^dagger O H - U1^dagger O H, whose first two and last two terms are identical, so D2 identically vanishes. Nevertheless, Eq. (B30) assigns D2 a nonzero contribution with coefficients 1/omega(-)_{nm} and 1/Omega(-)_{nm}, and Eq. (B32) keeps these terms, including diagonal n=m where omega(-)_{nn}=Omega(-)_{nn}=0 and the coefficients are singular. The text asserts these terms are negligible or vanish, but it never demonstrates the cancellation from the explicit expansion. Because the effective Hamiltonian is obtained by subtracting the anticommutator part from the second-order generator, a sign or operator-ordering error at this point would change H_eff, not merely a subleading correction. Every later result—the Rabi-block matrix Eq. (32), the beam-splitter operator Eq. (79), and the Hong-Ou-Mandel outputs Eqs. (94)-(95)—depends on this unverified factorization. A corrected derivation might preserve Eq. (11), but that has to be shown.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript develops a second-quantized effective-Hamiltonian description of a multi-Λ atomic gas coupled to two cavity modes, including center-of-mass motion. The central object is Eq. (11), H_eff = H_L + H_sp + H_pp, obtained by time-averaging the atomic field operators in Appendix B. From the reduced 2x2 Rabi-block matrix the authors derive a Raman beam-splitter operator, Eq. (79), and for a two-particle atomic input with equal photon numbers they predict both atomic and optical Hong-Ou-Mandel bunching, Eqs. (94)-(95). The paper is largely self-contained, with the derivation and all supporting commutator algebra placed in the appendices.","tokens_in":44685,"tokens_out":10713,"duration_ms":105500,"significance":"If the derivation is correct, the paper provides a useful parameter-free microscopic model that goes beyond the usual RWA treatment of Raman diffraction: it retains counter-rotating terms, includes contributions from all ancilla states, and identifies an effective particle-particle interaction induced by the quantized field. The Raman beam-splitter operator and the atomic/optical HOM predictions are concrete and falsifiable, and the appendix contains an explicit frequency hierarchy (Table B1) that makes the regime of validity testable. No fitted constants enter the effective Hamiltonian. However, the central derivation currently contains an algebraic error in Appendix B.1, so the significance is conditional on that step being repaired.","major_comments":[{"comment":"As printed, Eq. (B19) defines A† identically equal to A, and Eq. (B20) defines D2[O] as H O U1 − H O U1 + U1† O H − U1† O H, whose first two and last two terms cancel identically. Equation (B30) nevertheless assigns D2 a nonzero expression, and Eq. (B32) keeps those contributions, including the diagonal n=m terms in which the 1/ω(−)_nn and 1/Ω(−)_nn denominators are singular. The accompanying statement in Eq. (B20) that U1† is unitary is also incorrect for the first-order Dyson term in Eq. (B22). Since Eq. (B33) and hence Eq. (11) are obtained from these second-order expressions, the manuscript does not currently establish its central effective Hamiltonian; please supply corrected operator ordering and show explicitly how Eq. (B30) follows, because a sign or ordering error at this point changes H_eff rather than a subleading correction.","section":"Appendix B.1, Eqs. (B19)-(B20)"},{"comment":"The reduction to Eq. (11) relies on the ideal low-pass conditions (B24)-(B28), on the leading-order truncation that drops the 1/ω(−) and 1/Ω(−) terms in Eq. (B32), and on the neglected GHz-oscillating terms in V² in Eq. (C74). The text itself states in Appendix B.3.2 that a fully consistent calculation would require comparing the magnitudes of these contributions, but no such comparison is made, and no numerical benchmark of the averaged dynamics against the original Hamiltonian is provided. Because Eqs. (79), (94), and (95) all inherit these approximations, please add either an explicit analytic error estimate for the retained terms or a numerical check for the parameters of Table B1.","section":"Appendix B.3.2 and Appendix C"}],"minor_comments":[{"comment":"The two displayed definitions of A and A† are identical; even if this is only a typographical error, it should be corrected consistently with the subsequent derivation.","section":"Eq. (B19)"},{"comment":"The last line contains an extra equality sign: Ω(±)_BSvac = ∑ ω(±)_BSvac,j = ∑ = |Λ...|; the final '=' should be removed.","section":"Eq. (23)"},{"comment":"The phrase 'limit M→∞' is unclear because M is not defined in that section; please specify whether it is the atomic mass and how this limit removes the recoil phases.","section":"Section 4.3, after Eq. (95)"},{"comment":"Writing exp{±iω_n t} ≈ exp{±iΩ_n t} ≈ 0 is misleading: the two exponentials are not approximately equal to each other; presumably each time average vanishes separately under the low-pass filter.","section":"Eq. (B24)"},{"comment":"The denominator in the definition of Ω(R) is printed as Ω(+)_jj ab, whereas Eq. (44) uses ω(+)_jj ab; please reconcile this notation (there is also a lowercase n in the summation limit).","section":"Eq. (C71)"}],"recommendation":"major_revision","confidential_remarks":"The identified problems in Appendix B are load-bearing but appear fixable by a corrected derivation and an explicit consistency check; I do not see grounds for rejection if those repairs are made. My main concern is that the central appendix contains several sign/ordering typos in the very equations that produce Eq. (11), so the corrected version should be checked independently before acceptance."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"What you should know: this is a fresh second-quantized framework for multi-Lambda atoms in a two-mode cavity with center-of-mass motion, and it ends with a Raman-based Hong-Ou-Mandel prediction. The core idea is worth taking seriously, but the paper as printed has two concrete algebraic problems that need fixing before I would trust the general results.\n\nThe genuinely new piece is the operator-level time averaging applied to atomic field operators. The effective Hamiltonian in Eq. (11) keeps counter-rotating terms and generates an ancilla-mediated H_pp term. That is not in the cited semiclassical or first-quantized Raman literature, and the derivation has the right overall shape: no fitted parameters, and the HOM output follows from applying the derived unitary to a chosen state. The Rabi-block structure and the single-particle beam-splitter result reproduce known physics when the photon numbers are equal.\n\nNow the soft spots, in proportion. First, Eq. (B20) is not just a typo in a tangential appendix: D2 as printed is identically zero, while Eqs. (B30)-(B32) use it as nonzero with 1/omega(-) coefficients. The text also says U1-dagger is unitary, which is false for a first-order Dyson term. This is exactly the step that produces H_eff, so it has to be fixed; the final Hamiltonian might survive, but the current manuscript does not show it. The identical A and A-dagger in Eq. (B19) is part of the same problem.\n\nSecond, the general beam-splitter operator in Eq. (79) (and its rewriting in Eq. (81)) has swapped square-root factors for the two directions. Eq. (78), which comes from the same calculation, requires -i sin(theta|Omega|sqrt(n_a(n_b+1))) c / sqrt(n_a(n_b+1)) for a->b and the analogous expression with (n_a+1)n_b for b->a. Eq. (79) has the opposite pairing. For equal photon numbers the two factors coincide, which is why the HOM result in Eqs. (94)-(95) is unaffected. For general Fock states, the printed beam-splitter operator gives the wrong photon-number dependence.\n\nThe remaining concerns are minor by comparison: there is no numerical benchmarking of the delta-pulse and V^2 truncations, and H_pp, despite being advertised as a key finding, is not exercised in the applications. The timescale-separation assumptions in Table B1 are plausible for alkali Raman diffraction.\n\nWho should read this: people working on cavity QED with multi-level atoms and on quantized-light treatments of Raman diffraction. It deserves a serious referee. I would send it out, with the expectation of major revision focused on Appendix B and the beam-splitter operator.","headline":"A genuinely new second-quantized framework for multi-Lambda cavity QED with center-of-mass motion, but the printed averaging derivation has an internal inconsistency and the general beam-splitter operator has swapped square-root factors; both look repairable, and the equal-photon HOM result appears to survive.","tokens_in":45152,"tokens_out":22282,"would_cite":false,"duration_ms":234447,"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":"Time-averaging multi-Λ atomic field operators yields an effective Hamiltonian, counter-rotating terms included, that makes Raman diffraction a beam splitter for both atoms and light — giving atomic and optical Hong-Ou-Mandel bunching.","keywords":["light-matter interaction","Rabi model","multi-Λ system","second quantization","time averaging","effective Hamiltonian","beam splitter operator","Hong-Ou-Mandel effect"],"falsifier":"A numerical propagation of the unaveraged Hamiltonian of Eq. (10) — bosonic atoms with center-of-mass motion, two cavity modes, and the full ancilla manifold, using the alkali frequency scales of Table B1 — applied to the input $|\\psi^{(2)}\\rangle_A \\otimes |n\\,n\\rangle_L$ and read out after a pulse with $\\theta\\Omega_n = \\pi/4$ would settle the claim: any non-negligible amplitude in the $|a,b\\rangle$ or $|b,a\\rangle$ atomic channels, or in photon numbers other than $(n-1,n+1)$ and $(n+1,n-1)$, falsifies Eq. (94). An experiment preparing those two atoms and measuring their joint internal state plus the output photon numbers after the 50:50 Raman pulse is the direct laboratory version of the same test.","tokens_in":44166,"feed_emoji":"⚛️","tokens_out":22170,"duration_ms":195885,"temperature":0.7,"pith_summary":"This paper tries to establish that a bosonic multi-Λ atom in a two-mode cavity — a ground state and an excited state, each coupled to a manifold of ancillary states by one of two optical modes — can be described by an effective Hamiltonian obtained by time-averaging the atomic field operators rather than by eliminating levels. The effective Hamiltonian retains the counter-rotating terms that the rotating-wave approximation discards, collects AC-Stark and Bloch-Siegert shifts from the entire ancilla manifold, and generates a particle–particle interaction between atoms that is mediated by the ancilla states because the averaging sees the commutation relations of the quantized light field. From this Hamiltonian the authors derive a beam-splitter operator for Raman diffraction, and show that a two-atom input state with equal photon numbers in the two modes, after a 50:50 pulse, leaves the atoms and the light only in bunched configurations: both atoms in |b⟩ with the optical state |n−1,n+1⟩, or both in |a⟩ with |n+1,n−1⟩. That is the Hong-Ou-Mandel effect — twin inputs to a balanced beam splitter leaving together — predicted simultaneously for atoms and for light in one Raman configuration, giving a fully microscopic second-quantized model of a standard atom-optics pulse.","feed_headline":"Raman diffraction can bunch two atoms like photons","feed_subtitle":"Fully quantized Raman model with counter-rotating terms predicts atomic and optical Hong-Ou-Mandel interference.","key_machinery":"The carrying mechanism is a Hamiltonian averaging theory transferred from density-matrix dynamics (Ref. [67]) to operators: the time-averaged field operator evolves under an effective Hamiltonian assembled from second-order commutators of the co-rotating couplings $\\hat{h}_{j\\alpha}$ and counter-rotating couplings $\\hat{g}_{j\\alpha}$, evaluated with an ideal low-pass filter that keeps only frequency differences within a domain while all terahertz sums and cross-combinations average to zero (Eqs. (B24)–(B28)). The key structural step is that these commutators contain the light-field commutator $[\\hat{\\beta}^\\dagger, \\hat{\\alpha}]$ and delta-function terms from the atomic field operators, so the quantized nature of the field is what generates the ancilla-mediated particle–particle interaction $\\hat{H}_{\\rm pp}$, with couplings $\\zeta_{j\\alpha k}(R) = \\Omega_{j\\alpha}(R)/\\sqrt{\\omega^{(+)}_{jk\\alpha\\alpha}}$ and $\\eta_{j\\alpha k}(R) = \\Lambda_{j\\alpha}(R)/\\sqrt{\\Omega^{(+)}_{jk\\alpha\\alpha}}$ for the co- and counter-rotating channels. For the Raman application, a delta-pulse approximation and a decomposition of the interaction exponential into even and odd powers — using $\\hat{V}^{2k+1} = (\\hbar\\hat\\Omega)^{2k+1}(\\hbar\\hat\\Omega)^{-1}\\hat{V}$ — convert the evolution operator into the cosine–sine form of a beam splitter whose off-diagonals carry the momentum-transfer phases $e^{\\pm iKR}$.","core_discovery":"The paper's central claim is that the time-averaged evolution of the atomic field operator is governed by the effective Hamiltonian $\\hat{H}_{\\rm eff} = \\hat{H}_L + \\hat{H}_{\\rm sp} + \\hat{H}_{\\rm pp}$ (Eq. (11)) in the Schrödinger picture. $\\hat{H}_{\\rm sp}$ contains the single-particle physics: the AC-Stark and lowest-order Bloch-Siegert shifts, each summed over all $N-2$ ancilla states, together with ground–excited and ancilla–ancilla couplings, so that the ground and excited states decouple from the ancilla block at leading order and form a two-by-two Rabi-block matrix; the counter-rotating terms are kept, unlike in RWA treatments, and the ancilla states are not adiabatically eliminated. $\\hat{H}_{\\rm pp}$ is an induced two-body interaction, mediated by the ancilla states and produced by the commutation relations of the quantized field inside the averaging commutators, in a model that started with no atom–atom interactions. For the reduced Rabi-block Hamiltonian under a delta-pulse approximation, the paper evaluates the time-evolution operator on Gaussian single-particle states and identifies the Raman beam-splitter operator (Eq. (79)), which agrees with the semiclassical Raman result and with the Jaynes-Cummings beam-splitter form while carrying an extra optical–atomic phase that entangles the two sectors. For the two-particle input of Eq. (84) with $|n\\,n\\rangle_L$ light, the 50:50 pulse ($\\theta\\Omega_n = \\pi/4$) leaves only $|b,b\\rangle|n-1,n+1\\rangle$ and $|a,a\\rangle|n+1,n-1\\rangle$ (Eqs. (94)–(95)): the atomic and the optical Hong-Ou-Mandel effect in a single Raman configuration, and, in the atom-as-detector limit, an atom–light entangled version of the optical Hong-Ou-Mandel transformation.","pith_inferences":["The paper fixes the pulse at $\\theta\\Omega_n = \\pi/4$; a direct extension would scan the pulse area and predict a $\\sin^2(2\\theta\\Omega_n)$ Hong-Ou-Mandel curve — bunched output rising from zero to full as the pulse reaches 50:50 — so the model yields a continuous, testable interference signature, not just a single point.","The ancilla-mediated particle–particle interaction $\\hat{H}_{\\rm pp}$ has no RWA analogue in the paper's treatment; a natural testable consequence is a density-dependent shift or atom-pair correlation in a cavity that would show up only when the quantized nature of both matter and light is retained.","The authors note the framework transfers to magnetic-dipole transitions and, in principle, to ladder-type level schemes; if the averaging analysis carries over, each scheme would come with its own counter-rotating corrections and its own mediated interaction, giving a family of effective models for multi-mode cavity QED.","The beam-splitter phase $e^{-i\\theta C_{n_\\alpha,\\alpha}/\\hbar}$ entangles the optical and atomic sectors at large detuning, which suggests the formalism is a route to generating momentum entanglement in second quantization — the direction the paper's closing remarks point toward."],"forward_implications":["Raman diffraction gets a microscopic, second-quantized beam-splitter description: each pulse flips the internal state, swaps one photon between modes a and b, and transfers momentum $\\pm\\hbar K$, with the full ancilla-manifold AC-Stark and Bloch-Siegert shifts entering the diagonal phases.","A 50:50 Raman pulse on two atoms initially in different internal states produces only bunched outputs — two atoms in $|b\\rangle$ with $|n-1,n+1\\rangle$ light, or two atoms in $|a\\rangle$ with $|n+1,n-1\\rangle$ light — so the atomic Hong-Ou-Mandel effect is predicted in Raman diffraction, not only in Bragg diffraction.","With one photon per mode the same setup realizes the optical Hong-Ou-Mandel transformation, with the atoms serving as which-way detectors correlated to the photon output (Eq. (95)).","Because counter-rotating terms are kept, the effective Hamiltonian applies beyond the rotating-wave-approximation regime, and the Bloch-Siegert contributions are explicitly available with order-of-magnitude estimates for typical alkali-atom Raman parameters (Table B1).","The ancilla manifold is block-diagonalized away at leading order rather than eliminated, so ancilla dynamics — and the particle–particle interaction that reintroduces them — remain accessible in the effective theory, providing a hierarchy of approximations for multi-level cavity QED."],"supporting_citations":[{"why":"Supplies the time-averaging / effective-Hamiltonian formalism that the paper extends from density matrices to atomic field operators; the averaging machinery of the whole derivation rests on it.","marker":"[67]"},{"why":"The semiclassical theory of velocity-selective Raman transitions whose beam-splitter dynamics and Rabi oscillations the paper's operator result reproduces and extends.","marker":"[36]"},{"why":"Provides the Jaynes-Cummings beam-splitter operator and interaction-picture conventions against which Eq. (81) is benchmarked.","marker":"[74]"},{"why":"The paper's earlier treatment of Raman versus Bragg diffraction regimes; its operator techniques and delta-pulse setting underpin the Raman analysis of Section 4.","marker":"[37]"},{"why":"Defines the beam-splitter input–output relations that the two-particle Hong-Ou-Mandel analysis starts from.","marker":"[78]"},{"why":"The original two-photon Hong-Ou-Mandel experiment whose effect the paper extends to atoms and light in Raman diffraction.","marker":"[79]"},{"why":"The atomic Hong-Ou-Mandel experiment performed with Bragg diffraction, the precedent the Raman-configuration prediction extends.","marker":"[83]"},{"why":"Supplies the delta-pulse approximation used to convert the effective Hamiltonian into the time-evolution operator.","marker":"[69]"},{"why":"Defines the lowest-order Bloch-Siegert shift that the effective Hamiltonian retains beyond the rotating-wave approximation.","marker":"[28]"}],"fun_headline_variants":["Raman diffraction predicts atomic and optical HOM interference","Multi-photon Rabi model with COM motion shows photon bunching","Ancilla-mediated interactions yield atom-photon Hong-Ou-Mandel","Quantum Rabi framework captures Raman atom and light bunching","COM motion in cavity leads to atomic Hong-Ou-Mandel effect"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is a strict separation of timescales: the ideal low-pass averaging must make all sums and cross-combinations of the optical and atomic frequencies vanish while keeping only differences within each frequency domain, and the terms discarded afterwards — leading-order truncation of the detunings, gigahertz-oscillating pieces of $\\hat{V}^2$, and the assumption that the two atoms feel the same field across their separation — must stay negligible on the experimental timescale; if that hierarchy breaks, the effective Hamiltonian and the Hong-Ou-Mandel prediction change.","fun_headline_variants_meta":{"raw":{"variants":["Raman diffraction predicts atomic and optical HOM interference","Multi-photon Rabi model with COM motion shows photon bunching","Ancilla-mediated interactions yield atom-photon Hong-Ou-Mandel","Quantum Rabi framework captures Raman atom and light bunching","COM motion in cavity leads to atomic Hong-Ou-Mandel effect"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000277,"raw_usage":{"total_tokens":1802,"prompt_tokens":1250,"completion_tokens":552,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":866,"completion_tokens_details":{"reasoning_tokens":465}},"tokens_in":866,"tokens_out":552,"duration_ms":6056,"temperature":1.0,"reasoning_tokens":465,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T19:38:58.394835+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A numerical propagation of the unaveraged Hamiltonian of Eq. (10) — bosonic atoms with center-of-mass motion, two cavity modes, and the full ancilla manifold, using the alkali frequency scales of Table B1 — applied to the input $|\\psi^{(2)}\\rangle_A \\otimes |n\\,n\\rangle_L$ and read out after a pulse with $\\theta\\Omega_n = \\pi/4$ would settle the claim: any non-negligible amplitude in the $|a,b\\rangle$ or $|b,a\\rangle$ atomic channels, or in photon numbers other than $(n-1,n+1)$ and $(n+1,n-1)$, falsifies Eq. (94). An experiment preparing those two atoms and measuring their joint internal state plus the output photon numbers after the 50:50 Raman pulse is the direct laboratory version of the same test.","supporting_citations":[{"cited_title":"Phys Rev A 82:052106 53","cited_arxiv_id":null,"evidence_quote":"Supplies the time-averaging / effective-Hamiltonian formalism that the paper extends from density matrices to atomic field operators; the averaging machinery of the whole derivation rests on it."},{"cited_title":"Phys Rev A 45:342–348","cited_arxiv_id":null,"evidence_quote":"The semiclassical theory of velocity-selective Raman transitions whose beam-splitter dynamics and Rabi oscillations the paper's operator result reproduces and extends."},{"cited_title":"Wiley-VCH, Weinheim","cited_arxiv_id":null,"evidence_quote":"Provides the Jaynes-Cummings beam-splitter operator and interaction-picture conventions against which Eq. (81) is benchmarked."},{"cited_title":"Phys Rev A 101:053610","cited_arxiv_id":null,"evidence_quote":"The paper's earlier treatment of Raman versus Bragg diffraction regimes; its operator techniques and delta-pulse setting underpin the Raman analysis of Section 4."},{"cited_title":"Opt Comm 63(2):118–122","cited_arxiv_id":null,"evidence_quote":"Defines the beam-splitter input–output relations that the two-particle Hong-Ou-Mandel analysis starts from."},{"cited_title":"Phys Rev Lett 59:2044–2046","cited_arxiv_id":null,"evidence_quote":"The original two-photon Hong-Ou-Mandel experiment whose effect the paper extends to atoms and light in Raman diffraction."},{"cited_title":"Nature 520(7545):66–68 54","cited_arxiv_id":null,"evidence_quote":"The atomic Hong-Ou-Mandel experiment performed with Bragg diffraction, the precedent the Raman-configuration prediction extends."},{"cited_title":"New J Phys 15(1):013007","cited_arxiv_id":null,"evidence_quote":"Supplies the delta-pulse approximation used to convert the effective Hamiltonian into the time-evolution operator."}],"review_version":1}