{"id":"aa79f674-8009-4d94-acaa-91a4184dc3aa","arxiv_id":"2608.12481","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"The paper extends O(n) spectrum constraints to negative n via the O(n)-Sp(n) duality and derives closed-form two-loop anomalous dimensions for all phi^k operators from two known cases.","lead":"By treating the number of fields n in the O(n) model as a continuous variable, including negative even values, this paper derives constraints that force different operators to share the same scaling dimensions. The authors use these constraints to obtain a closed-form two-loop formula for the anomalous dimensions of every phi^k-type operator without computing new Feynman diagrams.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Two-loop phi^k formula hinges on the unproved claim that the four-index factorized diagram (4.5) has no 1/epsilon pole; this should be checked before the formula is accepted.","rationale":"The reader identified the weakest assumption as the claim that only f1 and f2 survive up to two loops, which rests on the pole structure of the factorized diagram (4.5). I agree this is the single most load-bearing concern: it is the essential step that turns a known two-loop result for two representations into a closed-form prediction for all representations. Without this step, the central formula could be incomplete. The paper cites Ref. [82] rather than proving the pole cancellation, and its own checks (Fig. 2) only demonstrate that spectrum constraints (i.e., coincidences at special n) hold, not that the absolute two-loop values match Eq. (4.10). Because the formula is new and extrapolates to arbitrary k, an explicit verification for at least one non-singlet representation is warranted. I do not see a more fundamental flaw: the spectrum-constraint machinery and the algebraic projection coefficients have been checked against explicit lower-order results, and the O(n)-Sp(n) duality is used only as an interpretation that is independently supported by the tensor-decomposition derivation of Sec. 3.3. Therefore, the appropriate verdict is the same conditional acceptance proposed by the reader, pending the concrete check described above.","tokens_in":27047,"tokens_out":25782,"duration_ms":232123,"concrete_test":"Extract from Refs. [43,44] the two-loop anomalous dimensions of the phi^8 operator in the T4 and T6 irreps (k=8, m=4 and m=2) and compare them with Eq. (4.7) using f1 and f2 from Eq. (4.10). If the O(lambda^2) coefficients do not match, diagram (4.5) contributes to the anomalous dimension and the formula fails for k=8, and by extension for general k. Alternatively, re-derive the 1/epsilon pole of diagram (4.5) for the phi^6 T4 operator using the algebraic factorization formula of Ref. [82, Sec.5] and show explicitly that the pole coefficient vanishes in MS.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central new result, Eq. (4.7)-(4.10), gives the two-loop scaling dimension of every phi^k-type operator in any traceless symmetric representation of O(n). The derivation assumes that the two-loop RG equation for the phi^k tensor contains only the f1 and f2 tensor structures of Eq. (4.2), i.e., that terms contracting four indices of c are absent at order lambda^2. The only two-loop topology that could generate such a term is diagram (4.5), which contracts four indices of c and factorizes into two one-loop subdiagrams on the middle vertex. The paper asserts, citing Ref. [82, Sec.5], that this diagram has a 1/epsilon^2 pole but no 1/epsilon pole in the MS scheme, so it does not contribute to anomalous dimensions. This step is load-bearing but is cited rather than proved in the text. If the diagram developed a 1/epsilon pole for some k, an f3 structure would appear, Eq. (4.7) would miss O(lambda^2) contributions, and the claimed new two-loop results would be incomplete. The paper's own consistency checks (e.g., Fig. 2) verify only the relative spectrum constraints, not the absolute values of f1 and f2 for intermediate representations. Footnote 12 further admits the argument can fail for momentum-dependent vertices, so it is not a trivial identity. Because the formula is a genuine extrapolation to arbitrary k and m, this unverified pole-structure claim is the most load-bearing assumption. The cited reference is likely correct, but the centrality of the claim warrants an explicit check.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper treats the number of flavors n in the O(n) model as a continuous variable and derives constraints on the operator spectrum when n is continued to negative even integers. The authors extend the quantum-evanescence spectrum constraints of Ref. [25] to negative even n using the O(n)/Sp(-n) duality, and complement this with an all-orders perturbative derivation in which the anomalous-dimension tensors are decomposed into the available O(n) tensor structures. The central new result is the two-loop master formula, Eqs. (4.7)-(4.10), giving the scaling dimension of any phi^k-type operator in an m-index traceless symmetric representation in terms of two functions f1 and f2 fixed from the known m=k and m=0 results. The paper also derives constraints on renormalization-group mixing matrices, predicts non-renormalization zeros in specific operator bases, and sketches extensions to OPE coefficients and to evanescence in the number of spacetime dimensions.","tokens_in":27409,"tokens_out":19294,"duration_ms":170990,"significance":"The master formula, if correct, is a valuable and nontrivial result: it extends two-loop anomalous dimensions to the entire family of phi^k operators without new diagrammatic computations, and it organizes infinite families of spectrum degeneracies at negative integer n. The perturbative re-derivation of the spectrum constraints in Sec. 3.3 is explicit and agrees with the five-loop and two-loop data shown in Figs. 1 and 2. The non-renormalization statements in Sec. 5, such as the one-loop zeros in Eq. (5.6), are concrete and testable. The main caveat is that the master formula rests on an unproved pole-structure assertion for diagram (4.5), so the result should be accompanied by a proof or an independent check before the formula can be fully accepted.","major_comments":[{"comment":"The entire two-loop master formula rests on the assertion that diagram (4.5), the only two-loop topology that contracts four indices of c_{phi^k}, has no 1/epsilon pole in the MS scheme and therefore does not generate an f3 tensor structure. This is cited to Ref. [82, Sec.5] but is not re-derived, and footnote 12 explicitly notes that the argument can fail when vertices carry momenta. Since f1 and f2 are subsequently fixed without using any intermediate-m two-loop datum, an unaccounted 1/epsilon pole would invalidate Eq. (4.7) for all 0<m<k. I request either a self-contained derivation of the pole structure for the phi^k vertex with generic k, or an explicit two-loop check of the final formula against the data of Refs. [43,44] for at least one intermediate representation.","section":"Sec. 4, Eqs. (4.2)-(4.7) and footnote 12"},{"comment":"The determination of f1 and f2 uses only the m=k and m=0 scaling dimensions, so agreement with those endpoints is automatic, and the spectrum constraints (4.8) follow from the algebraic form of (4.7) for any f1 and f2. The paper therefore provides no non-tautological check that the f2 extracted from the m=0 result also reproduces the two-loop dimensions of intermediate representations. I recommend adding a direct comparison with the two-loop results of Refs. [43,44] for a case such as k=6,m=2 or k=8,m=2, which would also provide an indirect check of the no-pole assumption.","section":"Sec. 4, Eqs. (4.7)-(4.10)"}],"minor_comments":[{"comment":"The displayed operator O_{phi^k,T_m} contains both (t.phi)^m and phi^{a1}...phi^{am}, which would give k+m fields rather than k; it should read O_{phi^k,T_m} = c_{phi^k,T_m} (t.phi)^m (phi.phi)^{(k-m)/2}, or an equivalent expression with traceless tensors. The subsequent formulas are consistent with the corrected form.","section":"Eq. (4.6)"},{"comment":"The symbols f1 and f2 are reused for OPE coefficients after being used for the renormalization-group functions in Eq. (4.2); please use different notation, such as g1 and g2, to avoid confusion.","section":"Sec. 6.1, Eq. (6.2)"},{"comment":"The two operators denoted c^{(S)}_{phi4□2,1} and c^{(S)}_{phi4□2,2} are not defined; please specify their tensor structures or the S4 Young diagrams to which they correspond.","section":"Eq. (5.4)"},{"comment":"The black dots in Fig. 1 are described as non-perturbatively constrained, but the plotted curves are perturbative results; please clarify that the constraints are exact statements satisfied order by order by these perturbative curves.","section":"Figs. 1 and 2 captions"},{"comment":"Since the extraction of f2 from the known two-loop m=0 result is a central step, please show the derivation explicitly rather than presenting the result without intermediate algebra.","section":"Eq. (4.10)"}],"recommendation":"major_revision","confidential_remarks":"The paper is within the journal's scope and the main result is worth publishing if the pole-structure issue is resolved. Note that the validation data in Figs. 1 and 2 come in part from Refs. [43,44,67], which are co-authored by one of the present authors; this is not improper, but it means the numerical validation is not fully independent."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: this paper earns its place. It takes the O(n)-Sp(n) duality at negative even n, makes it concrete with explicit tensor decompositions, and delivers a genuinely new closed-form two-loop anomalous dimension for every phi^k-type operator in O(n), Eqs. (4.7)-(4.10). I believe the formula is correct, but the step that makes it work is cited, not proved, and I would want it checked before treating the result as established.\n\nWhat is actually new: the extension of the Cao-Lu-Melia spectrum constraints to negative even n; the perturbative re-derivation of the constraints via the f-function parametrization, which makes the annihilation partners obvious; the two-loop formula itself; and the mixing-matrix factorization results in Sec. 5, including the n = -2 free-theory point. The paper is also honest in the right places: it states that the topology argument for diagram (4.5) fails when vertices carry momenta, and it uses that to explain why derivative operators do mix at two loops. That self-limitation makes me trust their reading of the literature.\n\nThe main soft spot is exactly what the stress test flags. The claim that diagram (4.5) has a 1/epsilon^2 pole but no 1/epsilon pole in MS is load-bearing for Eqs. (4.7)-(4.10), and it is deferred to Ref. [82, Sec. 5]. I think the cited result is right, but the stakes are high enough that an explicit check for one intermediate representation (phi^6 in T4, or phi^8 in T4) should be done or shown. Relatedly, the four-loop Sp(n) computation is mentioned in a footnote but not presented; that is a small credibility gap, not a flaw. And the figures mostly display the crossing structure, which is guaranteed by the representation theory regardless of the absolute values; the absolute two-loop numbers for intermediate m are interpolated from the m = k and m = 0 inputs, not independently verified. That is worth saying out loud, but it does not break the argument.\n\nThe citation pattern looks fine. The validation leans on the authors' own earlier results [43, 44, 67], but those are published and reproducible, so no issue.\n\nWho is this for: people computing anomalous dimensions in O(n) or general scalar EFTs, and anyone working on evanescence and non-renormalization. It deserves a serious referee. My recommendation: send it out, and have the referee check the factorization claim and ask to see the Sp(n) four-loop check.","headline":"New closed-form two-loop anomalous dimensions for all phi^k operators via O(n)-Sp(n) duality; real and useful, with one cited-but-unproved pole-structure claim that should be checked.","tokens_in":27915,"tokens_out":5354,"would_cite":true,"duration_ms":49631,"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":"The paper shows that continuing the number of flavors to negative even values forces degeneracies in the O(n) operator spectrum and yields a closed two-loop formula for the anomalous dimensions of all phi^k operators.","keywords":["O(n) model","negative number of flavors","spectrum constraints","Sp(n) duality","anomalous dimensions","phi^k operators","non-renormalization","epsilon expansion"],"falsifier":"Compute the two-loop $1/\\varepsilon$ pole of the factorized diagram (4.5) for $k=6$ in the minimal-subtraction scheme; a nonzero pole would directly invalidate the claim that only $f_1$ and $f_2$ contribute. Alternatively, an independent two-loop computation of $\\Delta_{\\phi^6,T_4}$ from standard graphs would expose any missing $O(\\lambda^2)$ piece.","tokens_in":26806,"feed_emoji":"📐","tokens_out":17601,"duration_ms":140880,"temperature":0.7,"pith_summary":"This paper treats the number of flavors $n$ in the $O(n)$ model as a continuous variable and shows that representation-theoretic degeneracies at special negative values of $n$ impose real constraints on the operator spectrum. The authors extend earlier spectrum constraints to negative even $n$ through the $O(2N)\\leftrightarrow Sp(2N)$ duality, which exchanges symmetrization and antisymmetrization of indices. In perturbation theory these constraints imply that, up to two loops, the renormalization of any pure $\\phi^k$ operator is governed by exactly two tensor structures. That reduces the two-loop anomalous dimension of every $\\phi^k$-type operator in every traceless symmetric representation to one closed formula, fixed by two previously known calculations. If correct, this predicts an infinite family of scaling dimensions without additional loop diagrams and explains factors of $(n\\pm a)$ in renormalization-group mixing matrices that become non-renormalization theorems.","feed_headline":"Formula gives two-loop dimensions of all phi^k operators in O(n)","feed_subtitle":"Negative flavor counts and the O(n)-Sp(n) duality turn two known results into an infinite family.","key_machinery":"The engine is the decomposition of reducible flavor tensors into $O(n)$ irreps, organized by Young tableaux and trace subtraction. The load-bearing identity is the dimension/character relation $d_\\lambda(n)=(-1)^m d_{\\lambda^{\\mathsf T}}(-n)$, which encodes the $O(2N)\\leftrightarrow Sp(2N)$ duality: when a continued character becomes the negative of another at some integer $n$, the two evanescent states must have equal scaling dimensions. On the perturbative side, the key simplification is that up to two loops the renormalization-group equation for a general $\\phi^k$ tensor contains only two structures, $f_1$ and $f_2$; the sole four-index-contracted two-loop diagram factorizes and carries no $1/\\varepsilon$ pole in the minimal-subtraction scheme, and mixing with derivative operators is absent at this order. This collapses an infinite family of anomalous dimensions to two unknown functions, which are then fixed by known one- and two-loop inputs.","core_discovery":"The central claim is a closed two-loop formula for the scaling dimension of any operator of the form $O_{\\phi^k,T_m}=(t\\cdot\\phi)^m\\phi^{a_1}\\cdots\\phi^{a_m}(\\phi\\cdot\\phi)^{(k-m)/2}$ in the $m$-index traceless symmetric representation of $O(n)$: $$\\Delta_{\\phi^k,T_m}=k(1-\\varepsilon/2)+f_1(\\$\\lambda$,k,n)+\\tfrac12(k-m)(n+k+m-2)f_2(\\$\\lambda$,k,n)+O(\\$lambda^{3}$),$$ with $$f_1(\\$\\lambda$,k,n)=\\frac{k(k-1)}3\\$\\lambda$-\\frac{k(2+$8k^{2}$+2k(n-6)-3n)}{36}\\$lambda^{2}$+O(\\$lambda^{3}$),\\qquad f_2(\\$\\lambda$,k,n)=\\frac{\\$\\lambda$}{3}-\\left(\\frac{k}{2}-\\frac79\\right)\\$lambda^{2}$+O(\\$lambda^{3}$).$$ The derivation rests on showing that only the two tensor structures multiplying $f_1$ and $f_2$ can appear in the renormalization-group equation up to two loops, after which the two functions are determined from existing one-loop and two-loop results rather than new calculations. The same degeneracies, continued to negative even $n$ through $O(2N)\\leftrightarrow Sp(2N)$ duality, give explicit spectrum constraints such as $\\Delta_{\\phi^6,S}=\\Delta_{\\phi^6,T_4}$ at $n=-2$ and, in general, $\\Delta_{\\phi^k,T_{m'}}=\\Delta_{\\phi^k,T_m}$ whenever $n=2-m-m'$.","pith_inferences":["The same two-input bootstrap could in principle be pushed to three loops once the third tensor structure and the mixing of $\\phi^k$ with derivative operators are computed for a single representative $k$; the paper identifies these as the missing ingredients.","Because all observed negative-$n$ degeneracies occur at even integers, odd negative values would require a supergroup extension such as $OSp$ models; the paper leaves this open, but the same annihilation logic would predict analogous constraints there.","A dedicated three-loop check of one nontrivial pair, for instance $\\Delta_{\\phi^6,S}$ versus $\\Delta_{\\phi^6,T_4}$ at $n=-2$, would test whether the spectrum constraints survive operator mixing; the paper expects all-loop validity but does not prove it.","The analogy with spacetime evanescence suggests that similar bootstrap relations could be obtained by treating the spacetime dimension $d$ as variable; the paper's large-$n$ example shows scalar and spin-two operators agreeing at $d=0$ through first order, hinting at a broader structure."],"forward_implications":["Every $\\phi^k$-type operator in any traceless symmetric representation of $O(n)$ receives a closed two-loop scaling dimension from Eqs. (4.7)--(4.10), so the whole infinite family is determined without new diagram computations.","At negative even $n$, spectrum constraints force degeneracies such as $\\Delta_{\\phi^6,S}=\\Delta_{\\phi^6,T_4}$ at $n=-2$ and $\\Delta_{\\phi^6,T}=\\Delta_{\\phi^6,T_6}$ at $n=-6$, and generally $\\Delta_{\\phi^k,T_m}=\\Delta_{\\phi^k,T_{m'}}$ when $n=2-m-m'$.","In a natural operator basis, singlet-sector mixing matrices acquire overall factors of $(n-a)$ and $(n+a)$; where such factors cannot be generated by one- or two-loop diagrams, the corresponding mixing entries vanish, yielding non-renormalization results at dimension six, eight, and ten.","At $n=-2$, the quartic interaction vanishes and the model is free; the associated $(n+2)$ factors in diagonal entries explain why some anomalous dimensions vanish there while evanescent operators can remain nontrivial.","The same annihilation constraints extend to operator product expansion coefficients, with the singlet and traceless-symmetric $\\phi^2$ OPE coefficients agreeing at $n=0$ in the known $\\varepsilon$-expansion data."],"supporting_citations":[{"why":"It establishes the spectrum-constraint mechanism: when an $O(n)$ character specializes to minus another at integer $n$, the two operators annihilate and their scaling dimensions coincide.","marker":"[25]"},{"why":"It provides the character identities and specialization rules used to continue $O(n)$ representations and their degeneracies.","marker":"[28]"},{"why":"It gives the dimension duality $d_\\lambda(n)=(-1)^m d_{\\lambda^{\\mathsf T}}(-n)$ linking orthogonal and symplectic representations.","marker":"[46]"},{"why":"It underlies the $O(-n)$/$Sp(n)$ correspondence through negative-dimensional tensors, the basis for extending constraints to negative even $n$.","marker":"[48]"},{"why":"It supplies the explicit $O(2N)$ to $Sp(2N)$ Lagrangian map with anticommuting scalars used to define and compute the negative-$n$ theory.","marker":"[72]"},{"why":"It collects the one-loop dimension formula for $\\phi^k$ operators and the two-loop singlet data that fix $f_1$ and $f_2$.","marker":"[29]"},{"why":"It provides the six-loop anomalous dimension of the maximally traceless-symmetric $\\phi^k$ operator, one of the two bootstrap inputs.","marker":"[84]"},{"why":"It provides the two-loop singlet anomalous dimension, the second bootstrap input.","marker":"[85]"},{"why":"It shows the factorized two-loop diagram (4.5) has no $1/\\varepsilon$ pole in the minimal-subtraction scheme, justifying the two-structure truncation.","marker":"[82]"},{"why":"It argues that $\\phi^k$-type operators do not mix with derivative operators up to two loops, limiting the tensor structures in the RG equation.","marker":"[81]"}],"fun_headline_variants":["Negative even n gives closed two-loop dimensions for all phi^k","O(n)-Sp(n) duality turns two loop results into infinite family","All phi^k scaling dimensions from a single duality trick","Non-renormalization and degeneracies from negative flavor count","Two-loop formula for every phi^k operator without new calculations"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing assumption is that up to two loops the renormalization of any pure $\\phi^k$ operator is captured entirely by two index-contraction patterns, meaning no mixing with derivative operators and no $1/\\varepsilon$ pole in the factorized four-index-contracted diagram, so that the two functions $f_1$ and $f_2$ suffice.","fun_headline_variants_meta":{"raw":{"variants":["Negative even n gives closed two-loop dimensions for all phi^k","O(n)-Sp(n) duality turns two loop results into infinite family","All phi^k scaling dimensions from a single duality trick","Non-renormalization and degeneracies from negative flavor count","Two-loop formula for every phi^k operator without new calculations"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000336,"raw_usage":{"total_tokens":1913,"prompt_tokens":1052,"completion_tokens":861,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":668,"completion_tokens_details":{"reasoning_tokens":776}},"tokens_in":668,"tokens_out":861,"duration_ms":8088,"temperature":1.0,"reasoning_tokens":776,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T00:09:51.934053+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the two-loop $1/\\varepsilon$ pole of the factorized diagram (4.5) for $k=6$ in the minimal-subtraction scheme; a nonzero pole would directly invalidate the claim that only $f_1$ and $f_2$ contribute. Alternatively, an independent two-loop computation of $\\Delta_{\\phi^6,T_4}$ from standard graphs would expose any missing $O(\\lambda^2)$ piece.","supporting_citations":[{"cited_title":"Constraints on the spectrum of field theories with non-integer $O(N)$ symmetry from quantum evanescence","cited_arxiv_id":"2312.10139","evidence_quote":"It establishes the spectrum-constraint mechanism: when an $O(n)$ character specializes to minus another at integer $n$, the two operators annihilate and their scaling dimensions coincide."}],"review_version":1}