{"id":"f38110ea-c687-4224-ad6c-8a9f6e7de5ec","arxiv_id":"2412.10618","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"In large N conformal field theories, states built from light scalar operators or stress tensors can have negative smeared null energy, but the negative amount is bounded by a scale set by the central charge.","lead":"The paper studies how negative the smeared null energy can be in strongly interacting quantum field theories with many degrees of freedom. It argues that, unlike free fields, these theories keep the negative energy bounded by a scale set by the central charge.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The p_max~N estimate and the 'at worst CT' scaling depend on an unverified assumption that connected odd-point correlators and multi-trace mixing can be tuned away; if this fails, the factorization regime shrinks to p~N^{2/3} and the claimed √CT enhancement is not established.","rationale":"The reader's weakest_assumption correctly identifies the factorization and mixing assumptions behind p_max~N. My analysis confirms this is the most load-bearing point: it supports the central scaling claims for multi-trace scalar states (√CT), the connected-correlator extension (C_T^{2/3} and beyond), and by analogy the multi-stress-tensor estimates. The paper itself flags the odd-point assumption in Section IV C, but the consequences are stronger than the text suggests: the two-three-point contribution (85) is not merely comparable to the four-point correction at p~N; it dominates for p>N^{2/3}. Whether this contribution can be removed by a subleading redefinition of the multi-trace operator is exactly the kind of technical step that would need to be checked before the factorization-based results can be accepted. I do not see an internal inconsistency or evidence of overclaiming: the authors label their strong conjecture as a conjecture and explicitly say the CT scaling has not been proven. However, the abstract's phrasing 'we present arguments that the negative smeared null energy ... scales at worst as CT' inherits the same fragility. Defenders of the paper could reasonably reply that the 'at worst' upper bound is safer when factorization breaks down, but the paper does not prove that; the breakdown could in principle enhance the negativity. The recommended verdict remains CONDITIONAL, so no change to the reader's verdict is needed. A concrete check, as proposed, would either restore confidence in the p~N estimate or force a weakened statement of the multi-trace scaling results.","tokens_in":34002,"tokens_out":11270,"duration_ms":121365,"concrete_test":"Recompute the norm and null-energy expectation value of |O_h^p> through O(N^{-2}) including the two-three-point contribution ⟨O_h†(O_h)^2⟩⟨(O_h)^2†O_h⟩ from Eq. (85), without imposing the odd-point vanishing assumption. If the ratio of this contribution to the leading p! term scales as O(p^3 N^{-2}), then factorization requires p<N^{2/3}, invalidating the p~N estimate in Eq. (86). To settle whether the assumption can be imposed consistently, construct the O(N^{-1}) correction to the state |[O^p]> that sets its overlap with |[O^{p-2}]> to zero and verify whether ⟨[O^p]|T[g]|[O^p]> is unchanged at relative order N^{-1}. If it changes, the estimates in Eqs. (87), (92), and (99) need revision.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The weakest load-bearing step is the p_max estimate in Section IV C. Equation (84) treats the connected four-point correction to the norm and gives a ratio O(p^2 N^{-2}), suggesting factorization holds up to p~N. But Equation (85) shows that two three-point functions contribute a ratio O(p(p-1)(p-2)N^{-2}) after dividing by p!, which is parametrically larger for p>N^{2/3}. The authors remove this by assuming, immediately after Eq. (85), that 'we have fixed all connected odd-point functions to vanish.' That fixing is not demonstrated at the required order: the integrated product (O_h)^2 has an O(N^{-1}) overlap with the single-trace O_h, and removing it requires subleading corrections to the operator/state [O^p] that can themselves change the numerator ⟨O^p† T-- O^p⟩ at the same order. If the assumption fails, the strict large-N factorization regime is p≪N^{2/3}, not p~N, so the √CT scaling of multi-trace states and the connected-correlator path to C_T^{a/(a+1)} in Eq. (99) are not established. More importantly, 'at worst CT' is an upper bound, but the paper only computes the leading factorized value up to the breakdown scale; nothing in the large-N expansion demonstrates that 1/N corrections beyond that scale cannot make ⟨T--⟩ even more negative. Thus the central claim is conditional on a monotonicity property of the 1/N expansion that is neither proven nor stated.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies how negative the smeared null energy can be in large N CFTs. It first shows that states prepared by scalar primary operators with conformal dimension Δ ≥ d have positive null energy density, and it gives evidence that pointwise stress-tensor insertions also give positive null energy in d ≥ 4. It then constructs states with negative smeared null energy using light scalars and superpositions involving the stress-energy tensor. The main claim is that, within the large N factorization approximation, the negative smeared null energy of states built from multi-trace operators scales at worst as the central charge C_T, with intermediate results scaling as √C_T, C_T^{2/3}, or C_T^{3/4}. The paper is explicit that these are arguments rather than proofs, and Section VII states that the bound proportional to C_T has not been proved.","tokens_in":34244,"tokens_out":4650,"duration_ms":50042,"significance":"If the central scaling claim can be made rigorous, the paper would be a significant step toward quantum energy inequalities in interacting CFTs in d > 2, extending known two-dimensional and free-field results. The paper has several genuine strengths: the positivity proof for scalar state norms in Appendix A is analytic; the conformal-frame reduction of the stress-tensor three-point function is a useful simplification; the product and holographic constructions provide concrete families of states with negative null energy of order C_T; and the authors are commendably explicit about which steps are conjectural. The main weakness is that the advertised \"at worst C_T\" statement is conditional on unproven large-N factorization and mixing assumptions, and the paper does not fully establish even the intermediate √C_T scaling once subleading corrections are tracked.","major_comments":[{"comment":"The estimate p_max ~ N is not established. Equation (85) gives a correction to the norm of order p(p-1)(p-2)N^{-2}, which is parametrically larger than the connected four-point correction of order p^2 N^{-2} in Eq. (84) once p > N^{2/3}. The authors remove this contribution by stating, immediately after Eq. (85), that they have fixed all connected odd-point functions to vanish, but they do not demonstrate that this fixing can be performed at the required subleading order without changing the numerator ⟨T--⟩ at the same order. Moreover, even the connected four-point correction in Eq. (84) is of order one relative to the leading term when p ~ N, so the regime of controlled factorization is p ≪ N, not p ~ N. The claimed √C_T scaling of multi-trace scalar states therefore needs either a proof of the odd-point/mixing assumption or a restriction to a smaller regime of p.","section":"Section IV C, Eqs. (84)-(87)"},{"comment":"The extrapolation to C_T^{a/(a+1)} is schematic. The argument resums connected correlators up to 2a_max, but the analysis in Appendix C establishes the asymptotic behavior of the ratio R only under assumptions about the dominance of particular combinatorial terms, and the text explicitly acknowledges in Section V A that terms where T-- itself appears in a higher-point connected correlator are ignored. Since the central claim is an upper bound — that negative null energy scales 'at worst' as C_T — the paper needs to show that 1/N corrections beyond the leading factorized value cannot make the null energy more negative in the regime where factorization breaks down. No such monotonicity statement is proved or even explicitly assumed.","section":"Section V A, Eqs. (92)-(99)"},{"comment":"The non-negativity of the stress-tensor three-point contribution in d ≥ 4 rests on numerical eigenvalue checks for free boson, fermion, and tensor structures, combined with the decomposition (62) and the Hofman-Maldacena constraints (63). The paper correctly acknowledges that this is not a proof. This issue is load-bearing for the stress-tensor state analysis: Section VI constructs negative null energy by using the assumed positivity of the ⟨TTT⟩ contribution to isolate the cross terms in the superposition (124). If a negative eigenvalue of B_{ab--cd} exists for some d ≥ 4, the scaling estimates in Eqs. (131)-(134) would need to be revisited.","section":"Section III C, Figure 1 and Table I"},{"comment":"The definition of the multi-trace state |O^p_h⟩ via iterated OPE projections is only valid to leading order in N. The paper does not verify that the subleading corrections required to define [O^p] as a primary operator, including the order N^{-1} mixing with the single-trace operator O, leave the leading factorized expectation value in Eq. (82) unchanged. If those corrections shift the numerator and denominator at the same order as the leading p-dependent terms, the linear p amplification in Eq. (83) is itself conditional. This is the same assumption that drives the p_max estimate, and it should be stated as an explicit technical assumption with a clear estimate of its error term.","section":"Section IV B, Eqs. (80)-(83)"}],"minor_comments":[{"comment":"The display of B_{ab--cd} is very hard to read: the row and column labels use dots as separators that do not align with the block structure. Reformatting the matrix with explicit block separators or listing the nonzero blocks would substantially improve readability.","section":"Section III B, matrix display"},{"comment":"The phrase 'part-in-parcel' should be 'part and parcel'.","section":"Section I, Introduction"},{"comment":"The definition of the scaled preparation function h'_Δ(x) ≡ λ^{d-Δ} h_Δ(λx) would benefit from a short explanation of how the integration measure and the support of h transform, since this is the key step leading to Eq. (29).","section":"Section II A, Eq. (28)"},{"comment":"The estimate N_grav ~ G_N p^2 / ℓ^{D-2} would be clearer if the authors stated the bulk-dimension conventions explicitly and commented on the régime in which the number of exchanged gravitons is small compared with p; this would help the reader see why Eq. (113) is weaker than full factorization.","section":"Section V C, Eq. (110)"}],"recommendation":"major_revision","confidential_remarks":"The paper is honest and tackles an important question, but the advertised central claim — that negative smeared null energy scales at worst as C_T — is conditional on unproven assumptions about connected odd-point functions and multi-trace mixing. The authors should either prove those assumptions at the required order or explicitly demote the central claim to a conjecture supported by examples. The paper fits the journal's scope and, with those gaps addressed, could become a valuable contribution."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: this is a serious and unusually honest large-N paper, and it deserves refereeing, but it is not a proof of the C_T bound. The reader's take is about right; the stress-test note lands on a real soft spot.\n\nWhat is genuinely new: the distinction between heavy (Delta >= d) scalar primaries, whose states have positive null energy density, and light scalars, which can give negative smeared null energy; the collinear-frame eigenvalue analysis of the stress-tensor three-point function with evidence for positivity in d >= 4; and the construction of multi-trace and multi-stress-tensor states whose negative null energy grows with the order p, up to sqrt(C_T) within full factorization and potentially C_T in product or holographic theories. The paper explicitly labels the strong conjecture as a conjecture and says in Section VII that the C_T scaling has not been proved. That honesty is a real virtue.\n\nThe soft spots are concentrated where the stress-test says they are. The p_max ~ N estimate in Section IV C assumes that connected odd-point correlators can be fixed to vanish and that mixing of [O^2] with the single-trace O can be tuned away at the required order. The combinatorics in Eq. (85) show that two three-point contributions can beat the two-point factorization for p > N^{2/3}, so without the odd-point assumption the strict factorization regime is smaller. More importantly, even if p_max ~ N holds, the paper only computes the leading factorized contribution; nothing in the large-N expansion rules out negative 1/N corrections beyond that scale. So the 'at worst C_T' claim is a conjecture supported by product CFT constructions and holographic estimates, not a demonstrated upper bound. I would ask a referee to push hard on this point.\n\nThe d >= 4 positivity of the TTT contribution is also numerical, checked for free scalar, fermion, and tensor theories, with no code or data shipped. It is acknowledged as not a proof, but if the authors want that claim to be load-bearing, they should provide artifacts or a proof. The citation pattern is fine; self-citations are relevant and not excessive.\n\nWho should read this: people working on quantum energy inequalities, holographic constraints, and CFT bootstrap bounds on energy flux. It is a useful paper even where conditional. Recommendation: send it to peer review, and ask the referees to focus on the p_max derivation and the status of the TTT positivity claim.","headline":"A serious, honestly-labeled large-N argument that smeared null energy in interacting CFTs may scale at worst with the central charge, but the headline C_T scaling is conditional on factorization assumptions that are not fully proven.","tokens_in":34846,"tokens_out":2211,"would_cite":true,"duration_ms":25301,"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":"In large N CFTs, the negative smeared null energy of scalar-primary and stress-tensor states scales at worst with the central charge, supporting a C_T-linear bound for interacting conformal field theories.","keywords":["quantum energy inequalities","null energy condition","smeared null energy","conformal field theory","large N factorization","central charge","stress-energy tensor","averaged null energy condition"],"falsifier":"Compute the smeared null energy beyond the large N expansion in a concrete large N CFT, such as the 3d O(N) model, for multi-trace states with p ~ $N^{{1+ε}}$; if the negative value exceeds a constant times C_T with that constant independent of N, the central claim is false. More directly, any interacting CFT in d ≥ 3 admitting a state whose integrated smeared null energy scales as C_T^β with β > 1 would refute the conjecture.","tokens_in":33728,"feed_emoji":"⚛️","tokens_out":9357,"duration_ms":81813,"temperature":0.7,"pith_summary":"The paper tests a long-standing question: can interacting quantum field theories in more than two spacetime dimensions have a lower bound on smeared null energy, or do they suffer the same unbounded violations as free fields? It argues that within the large N approximation, states built from scalar primary operators or from the stress-energy tensor itself have negative smeared null energy that scales at worst linearly with the central charge C_T. This matters because it would mean interacting CFTs are better behaved than free conformally coupled scalars, where piling up particles makes smeared null energy arbitrarily negative. If the suggested bound is right, a general state-independent quantum energy inequality proportional to C_T would hold for generic interacting CFTs, extending a known two-dimensional result to higher dimensions.","feed_headline":"Large-N CFTs cap negative null energy at central charge","feed_subtitle":"Interactions stop particle-piling from making smeared null energy unbounded; the floor is set by the central charge.","key_machinery":"The load-bearing mechanism is large N factorization: single-trace operators have two-point functions of order one, three-point functions of order $N^{{-1}}$, and connected four-point functions of order $N^{{-2}}$, so correlators of multi-trace operators [O^p] and [T^p] are dominated by disconnected two-point functions at large N. The central charge C_T, defined as the coefficient of the stress-energy tensor two-point function and set equal to $N^{2}$, turns the counting of p insertions into an energy scale: the combinatorics of connected higher-point functions stops the multi-trace order at p_max ~ N or at most p_max ~ $N^{{4/3}}$ with connected four-point functions, which translates into $C_T^{{1/2}}$, $C_T^{{2/3}}$, or C_T bounds. For stress-energy tensor states, the argument also uses the collinear frame of the ⟨TTT⟩ three-point function, whose eigenvalue matrix is fixed up to three CFT parameters and constrained by bounds derived from the averaged null energy condition; the matrix is found numerically to be non-negative in d ≥ 4, leaving superpositions of T^p states as the source of negativity.","core_discovery":"On the paper's own terms, the discovery is that the free-field mechanism for unbounded negative null energy—adding more particles to amplify a one-particle violation—fails in large N CFTs before it can beat the central charge. States prepared by scalar primaries with conformal dimension Δ ≥ d have positive null energy density, while light operators with Δ < d can produce negative smeared null energy; multi-trace versions of such states amplify it by the order p of the operator. Large N factorization limits how large p can be: keeping only factorized two-point functions gives p_max ~ N, hence negative smeared null energy of order $C_T^{{1/2}}$; allowing connected correlators, product constructions, or holographic weak interactions pushes the scaling up to order C_T but not beyond. For stress-energy tensor states, superpositions of consecutive multi-stress tensors reach $C_T^{{3/4}}$ within strict factorization and order C_T in holographic theories. The paper reads this as evidence for its strong conjecture: a state-independent bound on smeared null energy with a constant linear in C_T, with a weaker spectrum-dependent variant kept open.","pith_inferences":["Beyond the paper: the same large N counting should apply to other observables built from repeated single-trace insertions, suggesting that energy densities of conserved-charge currents or R-symmetry currents may obey similar C_T-scaled floors.","Beyond the paper: computing the first 1/N correction to the minimal smeared null energy in a specific model, such as the critical O(N) model, would show whether the bound approaches the free-field value from above or below; the conjecture predicts corrections that do not remove the C_T floor.","Beyond the paper: the collinear-frame positivity of the ⟨TTT⟩ matrix is numerical evidence; a symbolic proof that the relevant matrix is positive semidefinite for all d ≥ 4 would turn the paper's stress-tensor positivity result into a rigorous statement.","Beyond the paper: if the strong conjecture survives contact with explicit models, semiclassical singularity theorems could be run on CFT states with a universal central-charge constant, whereas the weak conjecture would instead tie the bound to a handful of light expectation values."],"forward_implications":["If the central claim is right, the free conformally coupled scalar's unbounded smeared null energy is an artifact of being non-interacting; any interacting large N CFT would have a finite, C_T-scaled floor.","Heavy scalar primary states (Δ ≥ d) and, for d ≥ 4, states made by pointwise stress-energy tensor insertions would have positive null energy density, so negative smeared null energy requires light operators or superpositions with the vacuum.","A state-independent lower bound linear in C_T would give the first quantum energy inequality for interacting CFTs in d > 2, matching the known two-dimensional bound and providing a universal constant for semiclassical applications.","Multi-stress-tensor superposition states offer a universal construction that needs no light scalar in the spectrum, reaching C_T^{3/4} within strict factorization and order C_T in holographic theories.","If the weak conjecture holds, effective descriptions of states with bounded null energy would be characterized by expectation values of finitely many light operators, giving clear diagnostics for which effective field theories have lower-bounded energy densities."],"supporting_citations":[{"why":"Defines the double smeared null energy condition for free minimally coupled scalars, the free-theory baseline the paper contrasts with.","marker":"[3]"},{"why":"Provides the interacting-field example where negative one-particle energy density coexists with a state-independent quantum energy inequality.","marker":"[6]"},{"why":"Establishes the two-dimensional CFT bound linear in central charge that the paper's conjecture seeks to generalize to higher dimensions.","marker":"[8]"},{"why":"Gives the conformally fixed two- and three-point functions and the central charge normalization used in every explicit computation.","marker":"[10]"},{"why":"Supplies the large N factorization and multi-trace operator framework that underlies the state constructions.","marker":"[12]"},{"why":"Supports the expectation in d=4 that smeared null energy is bounded by the central charge, which this paper extends.","marker":"[16]"},{"why":"Constructs free-theory states with negative smeared null energy that the multi-trace argument amplifies and caps.","marker":"[17]"},{"why":"Provides the proof of the averaged null energy condition used to restrict the coefficients of the stress-tensor three-point function.","marker":"[24]"},{"why":"Derives the conformal collider bounds that constrain the three-point function parameters in the eigenvalue analysis.","marker":"[25]"}],"fun_headline_variants":["Large N CFTs bound negative null energy by central charge","Interactions stop unbounded null energy in large N CFTs","Negative smeared null energy capped at C_T in large N","Null energy floor set by central charge in large N CFTs","Large N limits negative null energy to central charge order"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The bound depends on the assumption that the large N counting of connected correlation functions is complete, in particular that odd-point contributions and overlaps between many-particle and few-particle operator states stay negligible, so the number of operators p can be pushed to N.","fun_headline_variants_meta":{"raw":{"variants":["Large N CFTs bound negative null energy by central charge","Interactions stop unbounded null energy in large N CFTs","Negative smeared null energy capped at C_T in large N","Null energy floor set by central charge in large N CFTs","Large N limits negative null energy to central charge order"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000206,"raw_usage":{"total_tokens":1372,"prompt_tokens":895,"completion_tokens":477,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":511,"completion_tokens_details":{"reasoning_tokens":391}},"tokens_in":511,"tokens_out":477,"duration_ms":4288,"temperature":1.0,"reasoning_tokens":391,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T15:46:25.968405+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the smeared null energy beyond the large N expansion in a concrete large N CFT, such as the 3d O(N) model, for multi-trace states with p ~ $N^{{1+ε}}$; if the negative value exceeds a constant times C_T with that constant independent of N, the central claim is false. More directly, any interacting CFT in d ≥ 3 admitting a state whose integrated smeared null energy scales as C_T^β with β > 1 would refute the conjecture.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the double smeared null energy condition for free minimally coupled scalars, the free-theory baseline the paper contrasts with."},{"cited_title":"Nonpositivity of energy density in Quantized field theories,","cited_arxiv_id":null,"evidence_quote":"Provides the interacting-field example where negative one-particle energy density coexists with a state-independent quantum energy inequality."},{"cited_title":"Quantum energy inequalities in integrable models with several particle species and bound states","cited_arxiv_id":"2302.00063","evidence_quote":"Supplies the large N factorization and multi-trace operator framework that underlies the state constructions."},{"cited_title":"Einstein gravity from ANEC correlators","cited_arxiv_id":"1904.05892","evidence_quote":"Derives the conformal collider bounds that constrain the three-point function parameters in the eigenvalue analysis."}],"review_version":1}