Asymptotic Safety

These are equations, theorem steps, or proof moves where math specialists identified compressed or unverified derivations. They are shown once here as the canonical deduplicated list so the review stays auditable without repeating the same flags in every math specialist report.

• high
  Eq. (1.4.2) - The Wetterich/ERGE equation for the effective average action is stated rather than derived. It is the formal engine for the later beta-function computations.

  If wrong: If this equation or its adaptation to gauge/gravity systems were invalid, the ERGE-based beta functions and the main four-dimensional fixed-point evidence reviewed in Section 1.4 would be unsupported.

• high
  eq. (1.4.8) (one-loop higher-derivative gravity flow for \tilde Λ, \tilde G) - β-functions with q_*≈1.440 are quoted as obtained by setting other couplings to FP values; intermediate steps, dependence on regulator choice, and consistency of ‘setting to FP values’ are not demonstrated.

  If wrong: Would directly undermine the existence/location of the depicted non-Gaussian fixed point and associated statements about qualitative agreement across approaches.

• high
  Eq. (1.4.9)-(1.4.10) - The Einstein-Hilbert truncation beta functions are central and highly nontrivial rational expressions. They are stated without showing the Hessian decomposition, gauge-fixing details, cutoff insertion, trace evaluation, or projection onto \tilde\Lambda and \tilde G.

  If wrong: The displayed non-Gaussian fixed point at \tilde\Lambda=0.171, \tilde G=0.701 and its UV-attractive complex critical exponents would not be established; this would substantially weaken the paper's central claim of four-dimensional ERGE evidence for asymptotic safety.

• high
  eqs. (1.4.3)–(1.4.7) (1/N matter-dominated fixed point and a_i numbers) - Heat-kernel/Mellin-transform extraction of coefficients a_i^(n) and the conclusion that a_i^(n)=0 for n≥3 with optimized cutoff are stated without showing the computation or assumptions controlling truncation artifacts.

  If wrong: If these coefficients (or the vanishing for n≥3) are incorrect, then the claimed fixed point values (1.4.6–1.4.7) and the conclusion about relevance/irrelevance of infinitely many operators in this limit would not follow.

• high
  eqs. (1.4.9)–(1.4.10) (Einstein–Hilbert truncation beta functions) - Explicit rational expressions for β_{\tilde Λ}, β_{\tilde G} are given without derivation; dependence on gauge parameter/shape function and background choice is only qualitatively discussed.

  If wrong: If these β-functions are wrong or strongly scheme dependent, the fixed point at (\tilde Λ,\tilde G)≈(0.171,0.701), critical exponents, and the main ‘evidence for FP’ narrative would fail.

• medium
  Eq. (1.2.6)-(1.2.7) - The quotienting of coupling space by field redefinitions is asserted locally: given a field redefinition, new couplings are assumed to exist such that the effective action is unchanged, and an adapted coordinate system separating essential and redundant couplings is assumed. The local existence and regularity conditions are not proved.

  If wrong: The claim that only essential dimensionless couplings must reach a fixed point would require qualification; redundant directions might not be globally or smoothly removable in the way assumed.

• medium
  Eq. (1.3.3)-(1.3.5) - The gravitational scaling argument that k and G/Z_g cannot both be eliminated is central to the discussion of gravity's special RG structure. The argument is plausible, but compressed, and Eq. (1.3.4) contains a likely typographical/dimensional error in \tilde Z_g.

  If wrong: The conclusions that Newton's constant must satisfy a fixed-point condition in cutoff units, and that the cutoff in Planck units is bounded at the fixed point, would be undermined or would require rederivation.

• medium
  Eq. (1.4.1) - The epsilon-expansion beta function \beta_{\tilde G}=\epsilon\tilde G-q\tilde G^2 is quoted without derivation, including the value q=38/3 for pure gravity.

  If wrong: The early epsilon-expansion evidence for a non-Gaussian fixed point would fail, though the later ERGE-based evidence would remain separately assessable.

• medium
  eq. (1.4.1) (2+ε expansion beta function) - β_{\tilde G}=ε\tilde G−q\tilde G^2 and q=38/3 are quoted; no derivation nor discussion of scheme dependence or higher-loop corrections beyond a citation.

  If wrong: Would weaken one of the listed independent lines of evidence for a UV-attractive fixed point.

• medium
  eq. (1.4.2) (Wetterich ERGE) - ERGE is quoted without derivation; in gravity additional steps (gauge fixing, ghosts, background split Ward identities) are only referenced.

  If wrong: If the specific implementation is inconsistent, all downstream beta functions extracted from ERGE truncations would be unreliable.

• medium
  Eq. (1.4.3)-(1.4.7) - The large-N matter-loop beta-function coefficients are obtained by invoking the heat-kernel expansion and optimized cutoff, but the actual coefficient extraction is not shown. The fixed-point values for \tilde\Lambda_* and \tilde G_* are then given from these coefficients.

  If wrong: The large-N fixed-point construction and its claimed relevance/irrelevance classification would be unreliable; however, it is only one strand of the paper's evidence.

• medium
  Eq. (1.4.4) and the transition to beta functions (1.4.5). - The heat-kernel expansion Tr f(Delta) = sum_n Q_{2-n}(f) B_{2n}(Delta) is stated, and the coefficients Q_n(f) are described as 'given by Mellin transforms of f for n>0' with the note that the optimized cutoff R_k(z) = (k^2 - z) theta(k^2 - z) is used. The explicit evaluation of these transforms to obtain the numerical coefficients a_i^(n) (e.g., a^(0) = (n_S - 4 n_D + 2 n_M)/(32 pi^2)) is not shown. This is a compressed derivation step that is central to the large-N limit fixed-point results.

  If wrong: If the Mellin-transform evaluation is incorrect or if a different cutoff function yields different coefficients, the fixed-point coordinates for Lambda-tilde and G-tilde (and the claim that higher terms vanish at the FP) would change. The large-N evidence would be unreliable in its specific numerical predictions, though the structural existence of a fixed point might still hold for a different cutoff choice.

• medium
  Eq. (1.4.8) - The one-loop higher-derivative gravity beta functions for \tilde\Lambda and \tilde G, including the numerical constant q_*≈1.440, are presented after citing prior calculations, but the derivation and projection onto these couplings are not included.

  If wrong: The claimed qualitative agreement between higher-derivative gravity and the large-N flow would be weakened, and Figure 1's fixed-point analysis would not follow.

• medium
  Eq. (1.4.9)-(1.4.10) and Eq. (1.4.8) with Fig. 1.1-1.2. - The central beta functions for the Einstein-Hilbert truncation and for the higher-derivative four-coupling truncation are taken from cited references (Lauscher & Reuter 2002, Codello & Percacci 2006) without re-derivation. The ERGE-to-beta-function extraction procedure (via heat-kernel expansion and optimized cutoff) is compressed; the reader cannot verify the algebraic steps from the text alone.

  If wrong: If the claimed fixed-point coordinates (e.g., Lambda-tilde ~ 0.171, G-tilde ~ 0.701 for Einstein-Hilbert) or the eigenvalue structure are incorrect due to an error in the original derivation or a gauge/cutoff artifact, then the central claim of gravitational asymptotic safety at the non-Gaussian FP would lack its most concrete quantitative support. The qualitative consistency across multiple truncations and methods would still be suggestive, but the specific numbers and the claimed eigenvalue pattern would be unreliable.

• medium
  eq. (1.5.1) and link to spectral dimension/propagator ∼ p^{−4} - Scaling hg_{μν}i_k ∼ k^{−2} and the inference that the propagator changes from p^{−2} to p^{−4}, with Fourier transform ‘log’ behavior and ‘effective two dimensions’, are sketched and rely on identifying k∝p and on truncation-level equations of motion.

  If wrong: Would break the claimed quantitative/qualitative match to CDT spectral dimension results; core asymptotic safety claim could remain but this corroboration would not.

• medium
  eqs. (1.2.6)–(1.2.7) and discussion of essential vs redundant couplings - The existence of an adapted coordinate system splitting couplings into redundant/essential and the ability to fix redundant ones via field redefinitions is asserted locally; no conditions (e.g., regularity, absence of anomalies/Jacobians) are given.

  If wrong: If the split fails (globally or due to measure/Jacobian effects), the claim that only essential couplings need approach a fixed point could be compromised, affecting the predictivity argument tied to the UV critical surface.

• medium
  eqs. (1.3.4)–(1.3.5) - Rescaling between cutoff units and Planck units is presented as straightforward; details of how k transforms under field rescaling in the ERGE setup (background-field dependence, regulator transformation) are not shown.

  If wrong: Could affect the argument about whether k or G can be made redundant and the interpretation of ‘bounded cutoff in Planck units’.

• medium
  sec. 1.2: claim “beta functions automatically finite” - The statement that integrating out a shell k→k−δk produces no divergences and thus beta functions are automatically finite is asserted without specifying regulator properties and renormalization conditions.

  If wrong: If finiteness fails (e.g., due to improper regulator choice or gauge issues), the formal basis for using ERGE-defined beta functions as well-defined objects would weaken.

• medium
  sec. 1.3: derivation η_g* = d−2 and propagator ∼ p^{−d} - Uses η_g = ∂_t log Z_g = ∂_t log \tilde Z_g + d−2 and the heuristic propagator scaling p^{−2−η}. This ignores potential tensor/projector structure and gauge subtleties; derivation is sketched.

  If wrong: Would undermine the specific claim that the graviton propagator scales like p^{−d} at the fixed point and later links to ‘effective two-dimensional’ behavior.

• medium
  sec. 1.4: claim “In this truncation the UV critical surface is three dimensional” (f(R) up to order 6) - Dimensionality of the UV critical surface is quoted from literature without providing the stability matrix computation or eigenvalue count.

  If wrong: Would weaken the predictivity claim (finite number of relevant directions) in extended truncations.

• medium
  Section 1.4, f(R) polynomial truncation statement - The claim that the fixed point persists for polynomial f(R) truncations up to order six and that the UV critical surface is three-dimensional is stated without equations or derivation.

  If wrong: The robustness claim against enlarged truncations would be less supported, though the Einstein-Hilbert and other lower-order evidence would remain.

• low
  eq. (1.2.3) - Scaling relation for Γ_k is asserted by dimensional analysis, with unconventional canonical dimensions (dimensionless coordinates, metric as area). The precise exponent b^{d_A} and b^{d_i} factors depend on these conventions and are not derived.

  If wrong: Would alter the precise definition of dimensionless couplings/fields and hence the form of (1.2.5), but the general idea of using dimensionless variables would remain.

• low
  Eq. (1.2.3)-(1.2.5) - The dimensional scaling property and the resulting dimensionless beta-function formula are stated compactly. The transition from beta_i(g_j,k)=k^{d_i} a_i(\tilde g_j) to \tilde\beta_i=a_i-d_i\tilde g_i is plausible and standard, but the assumptions behind absence of explicit k-dependence beyond canonical dimensions are not fully spelled out.

  If wrong: The definition of fixed points in dimensionless coupling space would need modification, but the general asymptotic-safety framework could still be reformulated with explicit scale dependence.

• low
  eq. (1.3.2) and statement about inverses being the couplings (λ, ξ) - Identification of ‘physical’ couplings as inverses of g_i^(2) is asserted; mapping depends on normalization conventions of C^2 and R^2 terms.

  If wrong: Would mostly rescale later beta-function coefficients and the interpretation of asymptotic freedom for λ, ξ.

• low
  Eq. (1.4.1) -- epsilon-expansion beta function for pure gravity. - The beta function beta_G = epsilon G - q G^2 with q = 38/3 for pure gravity is cited from Weinberg (1979), Kawai & Ninomiya (1990), and Aida & Kitazawa (1997). No derivation is provided; the reader is simply told the result. The continuation from d=2+epsilon to d=4 (epsilon=2) is noted as questionable ('it was not clear whether one could trust the continuation of this result to four dimensions') -- this is a self-acknowledged gap, but the result is used as part of the cumulative evidence for a fixed point.

  If wrong: The epsilon-expansion is not the primary evidence for the 4D fixed point; it serves as historical context and qualitative support. If the two-loop q value were wrong or the continuation to epsilon=2 were invalid, the main 4D ERGE results would be unaffected. The paper itself acknowledges the uncertainty, so this is not a hidden gap.

• low
  Eq. (1.5.1) and subsequent propagator/spectral-dimension argument - The scaling of the effective metric in the fixed-point regime and the inference of a p^{-4} propagator/logarithmic short-distance behavior are sketched rather than fully derived.

  If wrong: The application to fractal spacetime and comparison with spectral dimension in causal dynamical triangulations would be weakened, but the core asymptotic-safety fixed-point argument would not depend on it.

• low
  eqs. (1.2.1)–(1.2.2) - Definition of W_k and Γ_k uses schematic functional measure and sign conventions; finiteness/normalization assumptions are implicit and not spelled out (e.g., existence of the modified Legendre transform for the chosen regulator).

  If wrong: Would mainly affect formal justification of Γ_k interpolation properties; downstream conceptual discussion of RG flow would need additional conditions.