{"id":"cf933e14-5028-4421-b746-96adecd06bfc","arxiv_id":"1908.06303","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"In a unitary CFT with a lowest-twist scalar primary, the scalar contributes to the lightcone OPE only at subleading order, and no positive averaged operator analogous to the ANEC exists for a scalar integral.","lead":"This paper derives the lightcone operator product expansion for a conformal field theory whose lowest-twist operator is a scalar, and argues that a scalar cannot have an averaged null energy condition analogue. The result separates the special role of the stress tensor from scalar operators in lightcone limits.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No-ANEC conclusion relies on completeness of the scalar OPE ansatz; the contour argument only rules out a pole for the specific O(σ) term.","rationale":"I agree with the reader's weakest_assumption that the no-ANEC conclusion is the least secure part of the paper. The derivation of the scalar lightcone OPE in Sec. 3 is a standard conformal three-point matching and appears internally consistent; the N=4 ordering argument (stress tensor prefactor v u^2 versus scalar prefactor v) is consistent with the coordinates used. The genuine soft spot is Sec. 4: the contour argument demonstrates that the specific O(σ) term in the scalar contribution produces no pole at σ=0, hence no positivity bound from that method, but it does not rule out other positivity derivations or establish a no-go theorem. This is a proof-gap concern rather than evidence that the conclusion is false, since the scalar light-ray operator is plausibly indefinite for unitary CFTs. The reader's CONDITIONAL verdict already captures this uncertainty, so no change to the verdict is warranted.","tokens_in":9268,"tokens_out":45052,"duration_ms":447884,"concrete_test":"Independently derive the scalar primary contribution to the lightcone OPE directly from the conformal partial wave expansion, without assuming the integral ansatz (3.3). In particular, compute the leading σ-behaviour of 1-G(σ) after the substitution v=-ησ, u=1/σ, starting from the sum over conformal descendants in (3.1). If conformal symmetry enforces that the coefficient of σ^{-1} vanishes identically for a lowest-twist scalar, the no-ANEC conclusion is robust; if any σ^{-1} component appears, the paper's central claim is unsupported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that no scalar analogue of ANEC exists is proven in Sec. 4 by showing that the scalar contribution to the lightcone OPE (3.8) is O(σ) in the coordinates v=-ησ, u=1/σ, so no pole at σ=0 appears and the contour integral yields no positivity bound. This conclusion inherits the integral ansatz (3.3)-(3.7): the OPE is written as 1 + λ(uv)^{Δ/2}/u ∫ du' K(u,u') φ(u',0), with K determined by matching the three-point function. If the true conformal block resummation of the scalar and its descendants contained terms with a different (u,v) scaling that survive in the lightcone limit, or if the scalar is not the unique lowest-twist operator (as in the N=4 example the paper itself discusses), then m-n+1 could vanish and a positive bound on a scalar light-ray operator could exist. The paper does not prove that the ansatz is exhaustive, nor that no other smearing of the scalar operator yields the desired pole. Thus the broad abstract statement 'there does not exist a scalar analog of the ANEC' is stronger than the contour argument establishes.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the lightcone OPE of two identical scalar primaries in CFTs where a scalar primary has the lowest twist. The authors propose an integral representation for the scalar contribution to the lightcone OPE, match its kernel against the three-point function, and apply the formalism to the long-range Ising model and to N=4 SYM. They argue that in N=4 SYM, where the stress tensor and a scalar share the lowest twist, the stress tensor dominates at leading order. The paper also attempts to rule out a scalar analogue of the ANEC by applying the causality contour argument of Hartman–Kundu–Tajdini to the derived scalar contribution.","tokens_in":9441,"tokens_out":21843,"duration_ms":210511,"significance":"The paper addresses a natural question in conformal field theory: what happens to the lightcone OPE and the ANEC when the lowest-twist operator is a scalar rather than the stress tensor? If the claims were correct, the result would clarify the universality of lightcone OPEs and delimit the ANEC program. The core three-point-function matching in Sec. 3 is standard and, as far as it goes, appears sound. However, the two headline conclusions—the N=4 leading-order statement and the non-existence of a scalar ANEC—are not supported by the paper's own equations. The N=4 claim conflicts with the scaling of the terms in Eq. (3.14), and the no-ANEC derivation is internally inconsistent with the stress-tensor proof reproduced in Appendix A. Because both claims are central to the abstract and conclusions, the manuscript in its current form is not publishable.","major_comments":[{"comment":"The conclusion that the stress tensor contributes at leading order in the lightcone OPE for N=4 SYM is inconsistent with the scaling of the terms written in Eq. (3.14). The stress-tensor term is proportional to \\(v/u^2\\), while the scalar term is proportional to \\(v\\). In the stated lightcone limit (|v|<<1, |u|>>1, |uv|<<1), the ratio of the scalar term to the stress-tensor term is \\(u^2\\), so the scalar term dominates. Thus Eq. (3.14) implies the opposite of the claim in the text and abstract. The argument that the kernels decouple in the three-point functions does not change this kinematic comparison.","section":"Sec. 3.2, Eq. (3.14)"},{"comment":"The contour calculation in Sec. 4 computes the \\(\\sigma\\)-dependence of \\(1-G(\\sigma)\\) solely from the prefactor \\(u^n v^m\\) in Eq. (4.2), treating the light-ray operator \\(S\\) as a \\(\\sigma\\)-independent local insertion. This is not a valid reduction for an operator defined by an integral over \\(u'\\), such as \\(S\\) in Eq. (3.8). The internal inconsistency becomes evident when the same logic is applied to the stress tensor: using the OPE (2.5) with \\(v/u^2\\), i.e., \\(n=-2, m=1\\), gives \\(m-n+1=4\\) in Eq. (4.9), predicting no ANEC for the stress tensor, which contradicts the known result. Appendix A itself uses Eq. (A.8) with a \\(1/\\sigma\\) pole, not the \\(\\sigma^3\\) behaviour that would follow from naive substitution of (2.5). The missing step is the treatment of the \\(u'\\) integration, which in the HKT proof is responsible for the pole. Therefore the derivation does not establish the absence of a scalar ANEC.","section":"Sec. 4 and Appendix A, Eqs. (4.6)–(4.10) and (A.8)"},{"comment":"The statement that 'there does not exist a scalar analog of the ANEC' is substantially stronger than what the analysis shows. Even if the contour argument of Sec. 4 were correct, it would at most show that the particular operator \\(S\\) defined in Eq. (3.8) does not yield a positivity bound from the specific causality argument of HKT. The paper does not rule out the existence of a positive scalar light-ray operator constructed with a different smearing, nor does it show that no other method could prove positivity. A claim of non-existence requires a broader argument, which is absent. The conclusion should be softened to a statement about the failure of this particular derivation.","section":"Abstract and Sec. 5"}],"minor_comments":[{"comment":"The notation 'vu3' in Eq. (2.3) is ambiguous; it should be written as \\(v/u^3\\) or with an explicit fraction to avoid confusion with \\(v u^3\\). The same ambiguity appears in the stress-tensor OPE terms.","section":"Eq. (2.3)"},{"comment":"The discussion of the long-range Ising model would benefit from stating explicitly that the \\(\\mathbb{Z}_2\\)-odd operator \\(\\phi\\) cannot appear in the \\(\\psi\\psi\\) OPE and that the lowest-twist even operator \\(\\phi^2\\) is the one actually exchanged; the current text introduces \\(T_{\\mu\\nu}\\) and \\(\\phi^2\\) but the logic of which operator dominates is not spelled out.","section":"Sec. 3.1"},{"comment":"The reproduction of the ANEC proof is terse. In particular, the step from Eq. (A.9) to the positivity of \\(\\langle O^\\dagger E O\\rangle\\) in Eq. (A.13) should explain how the rotation \\(R\\) maps the null line to itself and how the contour in the \\(\\sigma\\)-plane is preserved; without this explanation the proof is hard to follow.","section":"Appendix A"},{"comment":"The symbol for the OPE coefficient is not consistently typeset: it appears as \\(\\lambda_{\\psi\\psi\\phi}\\), \\(\\lambda_{\\psi\\psi O}\\), and \\(\\lambda_{\\psi\\psi T}\\) without a uniform convention for denoting the exchanged operator. Please standardise.","section":"Throughout"}],"recommendation":"reject","confidential_remarks":"The paper is clearly written and the three-point-function matching in Sec. 3 is competently done, but the two central claims fail on internal grounds. The N=4 ordering claim contradicts the scaling of Eq. (3.14), and the no-ANEC derivation is inconsistent with the known stress-tensor ANEC proof: the same logic applied to the stress tensor would predict no ANEC, which is false. These are not presentation issues but load-bearing errors that would require a substantive reworking of the arguments. I would encourage the authors to revisit the contour calculation with careful treatment of the \\(u'\\) integration and to scale back the abstract's claims accordingly."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThis paper does two things: it writes down the lightcone OPE for a scalar lowest-twist operator, with kernel K = (1 - u'^2/u^2)^{Δ/2-1}, and it argues that no scalar analogue of the ANEC exists. The OPE derivation in Sec. 3 is elementary conformal three-point matching and, as far as I can tell, correct. That is a useful extension of the HKT program, and the N=4 ordering example (stress tensor leading, scalar subleading) is a clean illustration.\n\nThe no-go in Sec. 4 is more subtle than the authors let on. They start from a general OPE of the form 1 + λ u^n v^m S, with n=Δ/2-1, m=Δ/2 for a scalar. In the coordinates v=-ησ, u=1/σ this gives a contribution ~ σ, not ~ 1/σ, so the contour integral has no residue and no positivity bound follows. The reader's worry about \"another smearing\" is, I think, not a real loophole: the u,v scaling is fixed by the three-point function, so any light-ray operator built from this OPE term would have the same σ dependence. The conclusion \"no scalar analogue of the ANEC\" is thus kinematically correct for the natural scalar light-ray operator.\n\nThe soft spots are real but minor. The abstract says \"a scalar is the lowest twist operator,\" while Sec. 4 needs it to be the unique lowest-twist scalar; that should be tightened. The long-range Ising example assumes φ^2 is a primary in the interacting theory, which is not checked and is questionable in a Wilson-Fisher-like fixed point. The N=4 discussion doesn't compute the relative OPE coefficients, so the leading/subleading claim is a scaling argument rather than a quantitative one. The proof also inherits the usual unitary-CFT caveats from the HKT analyticity and Rindler positivity, though the authors do flag that.\n\nOverall, this is a competent and honest short paper. It doesn't resolve the scalar-ANEC question in a profound way—the absence of a scalar ANEC is basically what you'd expect from the kinematics—but it makes the point cleanly and gives the kernel explicitly. I'd send it to a serious referee, with a request to qualify the abstract and check the Ising example. I'd probably cite it if I were writing about light-cone bounds.\n\nBest,\n[Your name]","headline":"A short, correct extension of the Hartman-Kundu-Tajdini lightcone OPE to scalar lowest-twist operators; the no-go for a scalar ANEC is kinematically right, but the abstract overstates its reach.","tokens_in":10049,"tokens_out":8236,"would_cite":true,"duration_ms":80946,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["81T40"],"pacs":["11.25.Hf"],"model":"deepseek-v4-flash","headline":"In a unitary CFT whose lowest-twist operator is a scalar, the scalar enters the lightcone OPE only at subleading order, and no scalar analogue of the averaged null energy condition can be derived.","keywords":["lightcone OPE","lowest twist operator","scalar primary","average null energy condition","conformal field theory","long-range Ising model","N=4 super-Yang-Mills","Rindler positivity"],"falsifier":"Compute the residue of $\\int d\\sigma\\,(1-G(\\sigma))$ for a scalar smeared with a different kernel $K'(u,u')$ that still matches the same conformal three-point function; if the residue is nonzero, the no-scalar-ANEC claim fails. More directly, find any unitary CFT with a scalar as the unique lowest-twist operator in which $\\langle O^\\dagger S O\\rangle$ is shown to have a definite sign.","tokens_in":9000,"feed_emoji":"⚛️","tokens_out":6850,"duration_ms":60982,"temperature":0.7,"pith_summary":"Two identical scalar operators in a conformal field theory usually see the stress tensor dominate their lightcone limit because it has the lowest twist. This paper asks what happens when the lowest-twist operator is itself a scalar, and claims that the scalar then enters only at subleading order, through a small prefactor $(uv)^{\\Delta_\\varphi/2}/u$ with a kernel fixed by conformal three-point functions. Because that suppression persists even when a scalar shares the lowest twist with the stress tensor, the stress tensor still sets the leading universal behaviour. From the same lightcone OPE the paper argues that no scalar analogue of the averaged null energy condition exists: the contour integral that proves positivity for the stress tensor evaluates to zero for the scalar. The examples are the long-range Ising model in $d=3$ and $\\mathcal{N}=4$ super-Yang-Mills theory in $d=4$.","feed_headline":"Scalar-only CFTs have no scalar null-energy bound","feed_subtitle":"The lightcone OPE puts lowest-twist scalars one order down, so the contour proof leaves nothing to bound.","key_machinery":"The central object is the integral-form lightcone OPE with an undetermined kernel $K(u,u')$ and prefactor $f(u,v)$. Matching the conformal three-point functions fixes the kernel to $K(u,u')=(1-u'^2/u^2)^{\\Delta_\\varphi/2-1}$ and the prefactor to $f(u,v)=(uv)^{\\Delta_\\varphi/2}/u$. The argument then transplants the contour proof used for the stress tensor: after the coordinate change $v=-\\eta\\sigma$, $u=1/\\sigma$, the scalar expectation value enters with $m-n+1=2$, making the relevant contour integral vanish identically and destroying any positivity bound.","core_discovery":"For a unitary CFT where a scalar primary is the unique lowest-twist operator, the lightcone OPE of two identical scalars takes the form $$\\frac{\\psi(u,v)\\psi(-u,-v)}{\\langle\\psi(u,v)\\psi(-u,-v)\\rangle} = 1 + \\lambda_{\\psi\\psi\\varphi}\\frac{(uv)^{\\Delta_\\varphi/2}}{u}\\int_{-\\infty}^{\\infty}du'\\left(1-\\frac{u'^2}{$u^{2}$}\\right)^{\\Delta_\\varphi/2-1}\\varphi(u',0),$$ so the scalar contribution is parametrically suppressed in the strict lightcone limit. If both the stress tensor and a scalar have minimal twist, as in $\\mathcal{N}=4$ super-Yang-Mills theory, the stress tensor still contributes at leading order and the scalar at subleading order. Repeating the causality-and-analyticity proof that yields the averaged null energy condition for the stress tensor produces a factor that vanishes when the scalar twist is inserted, so no sign bound on the smeared scalar operator can be extracted. The paper concludes that a scalar analogue of the averaged null energy condition does not exist in any unitary CFT where a scalar is the unique lowest-twist operator.","pith_inferences":["Beyond the paper, the same contour argument could be run for a conserved current as the lowest-twist operator; an 'average null charge' positivity condition would exist only if the current's twist produces $m-n+1=0$.","Beyond the paper, a numerical bootstrap or lattice simulation of the long-range Ising model could directly check the claimed subleading scalar correction by measuring the $u\\to\\infty$ behaviour of $\\langle\\psi\\psi\\varphi\\rangle$.","Beyond the paper, the result suggests that positivity conditions in CFTs are tied to the spin-2 twist family; lower-spin operators generically suppress the lightcone contribution and cancel the contour residue."],"forward_implications":["In any unitary CFT whose unique lowest-twist operator is a scalar, the lightcone OPE is dominated by the identity, and the scalar correction is parametrically suppressed by $(uv)^{\\Delta_\\varphi/2}$.","When a scalar shares the minimal twist with the stress tensor, the leading universal lightcone term is the stress tensor's averaged null energy operator; scalar operators act only at the next order.","The averaged null energy condition for the stress tensor continues to hold even when the stress tensor is not the lowest-twist operator, because the same contour proof goes through unchanged.","Non-unitary CFTs, such as the $\\phi^3$ Wilson-Fisher fixed point in $d=6-\\epsilon$ dimensions, fall outside the argument since Rindler reflection positivity is used to bound the correlator."],"supporting_citations":[{"why":"Supplies the integral lightcone OPE form and the contour proof of the averaged null energy condition that the paper transplants to the scalar case.","marker":"[10]"},{"why":"Establishes that operators of lowest twist dominate the lightcone OPE.","marker":"[2]"},{"why":"Fixes the three-point function normalisations used to determine the scalar kernel.","marker":"[30]"},{"why":"Provides the long-range Ising model in d=3 as the example of a unitary CFT with a scalar lowest-twist operator.","marker":"[20]"},{"why":"Shows that the leading spin-2 operator dimension in the long-range Ising model is such that the scalar has lower twist.","marker":"[24]"},{"why":"Supplies the causality and analyticity properties of correlators used in the contour argument.","marker":"[13]"},{"why":"Supplies wedge (Rindler) reflection positivity used to bound the correlator in the proof.","marker":"[32]"}],"fun_headline_variants":["No scalar ANEC when scalar is lowest twist","Lowest-twist scalar suppresses ANEC analog","Scalar lightcone OPE subleading, no null bound","CFT with scalar lowest twist lacks scalar ANEC","Why lowest-twist scalars have no ANEC analog"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the scalar's lightcone contribution is completely described by the kernel $(1-u'^2/u^2)^{\\Delta_\\varphi/2-1}$ with prefactor $(uv)^{\\Delta_\\varphi/2}/u$, and that the coordinate change $v=-\\eta\\sigma$, $u=1/\\sigma$ spans all legitimate ways to extract a positivity bound from the contour integral.","fun_headline_variants_meta":{"raw":{"variants":["No scalar ANEC when scalar is lowest twist","Lowest-twist scalar suppresses ANEC analog","Scalar lightcone OPE subleading, no null bound","CFT with scalar lowest twist lacks scalar ANEC","Why lowest-twist scalars have no ANEC analog"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000177,"raw_usage":{"total_tokens":1260,"prompt_tokens":881,"completion_tokens":379,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":497,"completion_tokens_details":{"reasoning_tokens":302}},"tokens_in":497,"tokens_out":379,"duration_ms":3788,"temperature":1.0,"reasoning_tokens":302,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T12:51:31.029448+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the residue of $\\int d\\sigma\\,(1-G(\\sigma))$ for a scalar smeared with a different kernel $K'(u,u')$ that still matches the same conformal three-point function; if the residue is nonzero, the no-scalar-ANEC claim fails. More directly, find any unitary CFT with a scalar as the unique lowest-twist operator in which $\\langle O^\\dagger S O\\rangle$ is shown to have a definite sign.","supporting_citations":[],"review_version":1}