{"id":"36090a90-b3aa-4977-825b-da7f3ddc6ba2","arxiv_id":"1908.02207","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"Uniform approximation of ideal photon subtraction and addition by fair beam splitter operations holds exactly under the stated energy moment conditions, and the conditions are tight.","lead":"The paper proves exactly when physically implementable photon subtraction and addition can be made to match the idealized versions: photon addition needs a state with bounded second energy moment, and photon subtraction additionally needs a state with energy bounded away from zero. It also supplies explicit error bounds and shows every relaxation of these conditions fails.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified: the uniform-convergence theorems and the necessity proofs are mathematically sound.","rationale":"The paper proves that the beam-splitter operations N_+(gamma) and N_-(gamma) converge uniformly on the stated moment-constrained classes, and that relaxing these constraints within the same moment hierarchy destroys uniform convergence. I checked the pure-state bounds, the purification step for mixed states, and the necessity constructions. The mixed-state extension is valid: the claimed inner-product lower bounds follow by monotonicity of e^{-x} and Cauchy-Schwarz, and the partial trace is contractive. The necessity examples use states whose second energy moment diverges (addition) or whose energy tends to zero (subtraction), so the moment constraints are genuinely needed for the specified classes. The only substantive caveat is that 'cannot be relaxed' should not be interpreted as a maximal-class statement: the set of Fock states has unbounded tr[rho H^2] yet both operations are realized exactly on it. However, the paper consistently discusses moment-constrained classes, and the abstract's wording refers to those conditions. The typos and the Prop. 2 limit-value error are cosmetic and do not affect the conclusions. Therefore the reader's CONDITIONAL verdict stands unchanged, with the caveat that the 'exact class' phrasing in the reader's summary is stronger than what the paper proves.","tokens_in":10973,"tokens_out":40069,"duration_ms":404401,"concrete_test":"Re-derive the purification inner-product bound in Prop. 5 using the correct operators |Phi> = sum_i sqrt(p_i) a |psi_i>|psi_i> / sqrt(F1) and |X> = sum_i sqrt(p_i) a e^{-gamma H} |psi_i>|psi_i> / sqrt(C); verify that <Phi|X> >= (F1 - gamma F2)/F1 and that trace-norm contraction gives the stated uniform bound. This directly settles the only under-derived step supporting the mixed-state theorem.","verdict_should_be":"UNCHANGED","load_bearing_attack":"After independent verification, I find no load-bearing flaw in the central claim. The mixed-state extensions in Props. 3 and 5, summarized as 'one can readily see,' are correct: for addition, the purification inner product satisfies <Phi|X> = tr[rho a e^{-gamma H} a-dagger] / sqrt(tr[rho a a-dagger] tr[rho a e^{-2 gamma H} a-dagger]) >= (tr[rho(H+1)] - gamma tr[rho(H+1)^2]) / tr[rho(H+1)], giving the claimed (F1+1 - gamma(1+2F1+F2))/(F1+1); for subtraction, the analogous computation gives >= (F1 - gamma F2)/F1. The necessity counterexamples in Props. 2 and 4 achieve the required lower bounds; the limit value in Prop. 2 should read 2 sqrt(E/(1+E)) rather than 2 sqrt(E)/(E+1), but this only strengthens the stated inequality. The plus/minus mislabellings in Prop. 5 are typos that do not affect the argument. If 'exact uniform-convergence class' is read as a maximal set, note that Fock states have unbounded tr[rho H^2] yet N_+ and N_- are exact on them; the paper's claim concerns the moment-constrained classes, not maximality. This is a caveat on interpretation, not a mathematical error.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the ideal photon subtraction and addition transformations ρ → t a ρ a† and ρ → t a† ρ a, which are not trace-nonincreasing and therefore are not valid quantum operations. The author considers physically motivated approximate operations N_-(γ) and N_+(γ) derived from beam-splitter models and asks when the conditional output states of these approximate operations converge uniformly, in trace norm, to the ideal normalized outputs. The main results are: (Prop. 3) for photon addition, uniform convergence holds on the class of states with bounded second energy moment tr[ρ H^2] ≤ E; (Prop. 4) for photon subtraction, this condition alone is not sufficient, because states with vanishing energy cause failure; and (Prop. 5) adding a lower bound tr[ρ H] ≥ E1 > 0 restores uniform convergence for subtraction. Proposition 2 shows that first-moment bounds alone are insufficient for either operation, and the paper also generalizes the results to multiple photon addition and subtraction. The proofs use explicit trace-norm bounds via inner products for pure states, purification for mixed states, and explicit counterexample families for the necessity claims.","tokens_in":11207,"tokens_out":14890,"duration_ms":139047,"significance":"If the results are correct, the paper provides a clean and essentially complete characterization of the moment constraints under which the standard approximate photon operations faithfully reproduce the ideal transformations in trace norm. This fills a genuine gap in the literature: previous work addressed approximations for individual states but not uniform convergence over state classes. The proofs are elementary, constructive, and parameter-free; the error bounds are explicit and the counterexamples are concrete. The paper also correctly identifies a qualitative asymmetry between photon addition and subtraction, namely the need for a nonvanishing energy lower bound in the subtraction case. The result is likely to be useful for quantum optics and for rigorous treatments of continuous-variable quantum operations. The main weaknesses are presentational: several typographical errors, a terse mixed-state step, and a misstated limit value in Prop. 2; none of these appears to affect the validity of the central conclusions.","major_comments":[],"minor_comments":[{"comment":"The displayed limit in the proof of Prop. 2 should read 2 sqrt(E/(E+1)) rather than 2 sqrt(E)/(E+1); as printed, the stated limit is smaller than the target bound sqrt(E/(E+1)) for E > 4, so the proof of inequality (8) is incomplete unless the correct value is used.","section":"Section III, proof of Prop. 2"},{"comment":"The step 'One can readily see that ⟨Φ|X⟩ ≥ (F1+1-γ(1+2F1+F2))/(F1+1)' is an omitted derivation that is load-bearing for extending the theorem to mixed states; please include the short argument, for instance by applying the pure-state bound to each spectral component and using Cauchy-Schwarz for the denominator.","section":"Section IV, Prop. 3, mixed-state paragraph"},{"comment":"There are copy-paste errors in this paragraph: the states should be the approximate and ideal subtraction outputs, not ~ρ_+ and N_+, and the denominator in the definition of |X⟩ should involve e^{-2γ a†a} (equivalently e^{-2γH}) rather than e^{-2γ aa†}.","section":"Section V, Prop. 5, mixed-state paragraph"},{"comment":"The sentence 'there exists s1 > 2 such that [expression] = ...' is imprecise because the expression depends continuously on s and only tends to the displayed limit as s → 2+0; the argument should say that the expression can be made arbitrarily close to that limit, hence larger than (e-1)/(2e).","section":"Section II, proof of Prop. 1, first case"},{"comment":"The phrase 'these conditions cannot be relaxed' should be understood relative to the moment-constrained families considered here; Fock states have unbounded tr[ρH^2] yet the approximate operations are exact on them, so the statement is not one of absolute maximality of the state set.","section":"Abstract and Introduction"},{"comment":"There are several typographical errors that should be corrected: 'nonicreasing' → 'nonincreasing', 'Simirlary' → 'Similarly', 'transmittence' → 'transmittance', and the title page has 'addi tion'.","section":"Throughout"}],"recommendation":"minor_revision","confidential_remarks":"The paper is mathematically sound in its central claims. The main issues are presentation and a few local errors that are straightforward to fix. The 'one can readily see' purification steps in Props. 3 and 5 should be expanded for the benefit of readers, and the limit-value typo in Prop. 2 should be corrected. I do not see any concern about novelty or citation practice; the result is a modest but solid contribution to the mathematical foundations of approximate photon operations."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe main result is solid and genuinely new. Filippov shows that for photon addition, uniform convergence of the approximate beam-splitter operation to the ideal map holds exactly over states with bounded tr[rho H^2]; for photon subtraction one additionally needs a lower bound tr[rho H] >= E1 > 0. The necessity proofs are explicit counterexamples, the sufficient direction gives simple error bounds in terms of gamma, E1, E2, and the multi-photon extension follows from a scaling relation. There are no fitted parameters, no numerics, and the cited operations are standard models. The paper deserves a serious referee.\n\nWhat I liked: Prop 2's counterexample is the right one; it shows first-moment constraints fail because high-number tails with fixed energy have divergent second moment. The second-moment bound in Prop 3 is the actual insight: using e^{-gamma(n+1)} >= 1 - gamma(n+1) and bounding 1+2F1+F2 is enough. The mixed-state purification step is only summarized as \"one can readily see,\" but the stress-test verification shows it is correct; the details should be written out in a revision, since that is the only non-elementary step.\n\nSoft spots, in proportion: the plus/minus labels in Prop 5's mixed-state paragraph are swapped (the text says rho_+ and N_+ where it means rho_- and N_-). Cosmetic but confusing. The gamma choice epsilon^2/[8(3E+2)] in Prop 3 is conservative but fine, not an error. The phrase \"cannot be relaxed\" should be read as \"cannot be relaxed within these moment-constrained families\": Fock states have unbounded tr[rho H^2] and are exactly mapped by both operations, so the classes are not maximal. That is an interpretation caveat, not a mathematical flaw. The citation pattern is appropriate, with due credit to the energy-constrained frameworks of Shirokov and Winter.\n\nWho is this for? Quantum optics and continuous-variable quantum information readers, especially experimentalists who want a concrete gamma for a target epsilon. The bounds are loose but explicit and correct.\n\nRecommendation: send to a competent referee. I would accept, conditional on a short revision that fixes the Prop 5 labels and expands the mixed-state proof from \"one can readily see\" to a few displayed inequalities.","headline":"A clean, mostly self-contained paper that pins down exactly when beam-splitter photon subtraction/addition converges uniformly to the ideal maps; worth refereeing after fixing typos and expanding the mixed-state proofs.","tokens_in":11711,"tokens_out":3136,"would_cite":true,"duration_ms":31607,"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":"Beam-splitter photon addition and subtraction are uniformly valid on two energy-moment classes, and these classes are optimal.","keywords":["photon subtraction","photon addition","quantum operation","energy constraint","uniform convergence","trace distance","beam splitter","energy moments"],"falsifier":"Compute, for fixed $\\gamma,E$ and the family $|\\psi_N\\rangle=\\sqrt{1-E/N^2}|0\\rangle+\\sqrt{E/(2N^2)}|1\\rangle+\\sqrt{E/(2N^2)}|N\\rangle$ from Proposition 4, the trace norm $\\|\\cdot\\|_1$ between the normalized ideal subtraction output and the normalized $\\mathcal{N}_-(\\gamma)$ output as $N \\to \\infty$. The paper predicts it approaches 2; if it stays below 1, the claimed impossibility fails. For photon addition, a single state in $\\mathcal{S}^{(2)}_{E_2}$ exceeding the stated $\\sqrt{8\\gamma(3E_2+1)}$ bound would disprove the uniform estimate.","tokens_in":10748,"feed_emoji":"⚛️","tokens_out":11118,"duration_ms":111156,"temperature":0.7,"pith_summary":"The paper asks when the textbook photon subtraction and addition maps, $\\varrho \\mapsto t a \\varrho a^\\dagger$ and $\\varrho \\mapsto t a^\\dagger \\varrho a$, can be replaced by real quantum operations. These maps are completely positive but not trace nonincreasing, so no experiment can implement them exactly; a beam splitter provides fair conditional operations $\\mathcal{N}_-$ and $\\mathcal{N}_+$ that approximate them. The main result is a sharp certificate for uniform approximation: photon addition converges on states with finite second energy moment $\\operatorname{tr}[\\varrho H^2] \\leq E_2$, while photon subtraction converges on states that also have energy bounded below by $E_1 > 0$. Both restrictions are proved necessary, and the result extends to multiple-photon subtraction and addition. This matters because it tells a user of a fixed beam-splitter setup exactly which input states can be trusted to behave like the idealized operation.","feed_headline":"Photon addition and subtraction need exact energy constraints","feed_subtitle":"Uniform convergence needs finite second energy moment for addition, plus nonzero energy for subtraction.","key_machinery":"The load-bearing objects are the two fair operations of Eqs. (4) and (5), beam-splitter conditional maps whose exponential damping $e^{-\\gamma H}$ with $H=a^\\dagger a$ makes the maps trace nonincreasing by cutting off high-Fock-number tails. The proof shows that on the relevant moment classes the damping error is controlled by simple inequalities of the form $\\sqrt{8\\gamma E_2/E_1}$ for subtraction and $\\sqrt{8\\gamma(3E_2+1)}$ for addition, so choosing $\\gamma$ small enough gives uniform $\\varepsilon$-closeness. A purification step then lifts the pure-state estimates to mixed states by contractivity of the trace distance under partial trace.","core_discovery":"The central discovery is that the physically realisable approximate operations $\\mathcal{N}_+(\\gamma)[\\varrho]=(e^{2\\gamma}-1)e^{-\\gamma a^\\dagger a}a^\\dagger \\varrho a e^{-\\gamma a^\\dagger a}$ and $\\mathcal{N}_-(\\gamma)[\\varrho]=(e^{2\\gamma}-1)a e^{-\\gamma a^\\dagger a}\\varrho e^{-\\gamma a^\\dagger a}a^\\dagger$, after normalization, converge uniformly to the ideal conditional outputs $a^\\dagger \\varrho a/\\operatorname{tr}[a^\\dagger \\varrho a]$ and $a\\varrho a^\\dagger/\\operatorname{tr}[a\\varrho a^\\dagger]$ exactly on the stated energy-moment classes. For addition, the class is $\\mathcal{S}^{(2)}_{E_2}(\\mathcal{H})$, states with $\\operatorname{tr}[\\varrho H^2] \\leq E_2 < \\infty$; for subtraction, the class is $\\mathcal{S}^{(1;2)}_{E_1;E_2}(\\mathcal{H})$, states with $\\operatorname{tr}[\\varrho H] \\geq E_1 > 0$ and $\\operatorname{tr}[\\varrho H^2] \\leq E_2 < \\infty$. The paper proves that finite energy alone cannot give uniform convergence for either operation, and that the second-moment bound alone cannot give it for subtraction, because families of states with energy tending to zero force the trace distance to its maximum possible value. The same mechanism, applied to $k$-fold operations, replaces the second moment by the $(k+1)$-th energy moment for multiple photon subtraction and addition.","pith_inferences":["The proof structure suggests the same uniform-convergence certificate should hold for any damping profile $e^{-f(H)}$ with $f$ growing at least linearly, not only the beam-splitter form; replacing $\\gamma n$ by $f(n)$ in the estimates would yield analogous moment conditions.","The sharp failure at vanishing energy points to a resource-theoretic reading: ideal photon subtraction is faithfully implementable exactly when the state carries at least one photon on average, so mean photon number acts as a necessary resource for the operation.","A direct experimental check could use the families in Propositions 2 and 4 as adversarial inputs: if a fixed setup is claimed to realize approximate subtraction over a broad class, measuring the trace distance on those families should reveal the divergence the paper predicts."],"forward_implications":["For any beam-splitter transmittance, a user can certify photon addition on all states with $\\operatorname{tr}[\\varrho H^2] \\leq E_2$ by choosing $\\gamma$ below an explicit threshold; no state-by-state verification is needed.","Photon subtraction cannot be certified on any class containing states of arbitrarily small mean energy, so experiments with very low photon number should not treat the beam-splitter output as ideal subtraction.","Multiphoton subtraction and addition inherit the same dichotomy: the $(k+1)$-th energy moment controls $k$-photon addition, while $k$-photon subtraction additionally needs a positive lower bound on energy.","The impossibility results give explicit stress-test states: the families in Propositions 2 and 4 are ready-made worst cases for any approximate implementation."],"supporting_citations":[{"why":"Supplies the beam-splitter realization of photon subtraction and addition, including the damping and prefactor in Eqs. (4) and (5) that define the fair approximate operations.","marker":"[13]"},{"why":"Provides the contractivity property of the trace distance under partial trace (Theorem 9.2), which extends the pure-state error estimates to mixed states in the proofs of Propositions 3 and 5.","marker":"[31]"},{"why":"Sets the energy-constrained framework and the state classes with bounded energy moments in which uniform convergence is formulated.","marker":"[25]"}],"fun_headline_variants":["Uniform convergence needs finite second energy moment for photon addition","Photon subtraction requires nonzero energy and finite second moment","Exact energy constraints proven for photon addition and subtraction","Photon operations: uniform convergence tied to energy moments","Optimal energy-moment bounds for photon op approximations"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The result assumes the beam-splitter operations (4) and (5) are the faithful physical approximations of photon subtraction and addition, with no additional loss, detector inefficiency, or noise; the purification bound stated as 'one can readily see' in Propositions 3 and 5 is also used without full derivation.","fun_headline_variants_meta":{"raw":{"variants":["Uniform convergence needs finite second energy moment for photon addition","Photon subtraction requires nonzero energy and finite second moment","Exact energy constraints proven for photon addition and subtraction","Photon operations: uniform convergence tied to energy moments","Optimal energy-moment bounds for photon op approximations"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000269,"raw_usage":{"total_tokens":1727,"prompt_tokens":1158,"completion_tokens":569,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":774,"completion_tokens_details":{"reasoning_tokens":493}},"tokens_in":774,"tokens_out":569,"duration_ms":5719,"temperature":1.0,"reasoning_tokens":493,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T14:51:06.814966+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute, for fixed $\\gamma,E$ and the family $|\\psi_N\\rangle=\\sqrt{1-E/N^2}|0\\rangle+\\sqrt{E/(2N^2)}|1\\rangle+\\sqrt{E/(2N^2)}|N\\rangle$ from Proposition 4, the trace norm $\\|\\cdot\\|_1$ between the normalized ideal subtraction output and the normalized $\\mathcal{N}_-(\\gamma)$ output as $N \\to \\infty$. The paper predicts it approaches 2; if it stays below 1, the claimed impossibility fails. For photon addition, a single state in $\\mathcal{S}^{(2)}_{E_2}$ exceeding the stated $\\sqrt{8\\gamma(3E_2+1)}$ bound would disprove the uniform estimate.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the beam-splitter realization of photon subtraction and addition, including the damping and prefactor in Eqs. (4) and (5) that define the fair approximate operations."},{"cited_title":"Multiphoton subtracted thermal states: De- scription, preparation, and reconstruction,","cited_arxiv_id":null,"evidence_quote":"Provides the contractivity property of the trace distance under partial trace (Theorem 9.2), which extends the pure-state error estimates to mixed states in the proofs of Propositions 3 and 5."},{"cited_title":"Smooth quantum- classical transition in photon subtraction and addition processes,","cited_arxiv_id":null,"evidence_quote":"Sets the energy-constrained framework and the state classes with bounded energy moments in which uniform convergence is formulated."}],"review_version":1}