Nojiri-Odintsov Conformal Anomaly Mechanism

• The Nojiri–Odintsov conformal anomaly mechanism is a framework where the four-dimensional trace anomaly of conformal matter generates effective nonlocal gravitational actions.
• It employs key curvature invariants (C², E₄) and the Paneitz operator (Δ₄) to encapsulate the trace anomaly effects, impacting semiclassical gravity and cosmology.
• The mechanism links anomaly poles from renormalization with long-range infrared effects, offering insights into resolving cosmological singularities and modifying gravitational dynamics.

The Nojiri–Odintsov conformal anomaly mechanism denotes the use of the four-dimensional trace anomaly of conformal matter as an effective source for semiclassical gravitational dynamics, typically through an anomaly-induced effective action that is either nonlocal or rewritten in local auxiliary-field form. In the standard formulation, the mechanism is organized by the curvature invariants $C^2$, $E_4$, and $\Box R$, and by the conformally covariant fourth-order operator $\Delta_4$; in cosmological applications it is used to modify gravitational evolution, including inflationary, late-time, or singularity-resolution scenarios (Lucat et al., 2017, Barvinsky et al., 2023).

1. Core geometric and functional structure

In four dimensions, the trace anomaly is written in the standard curvature basis

$$\langle T^\mu{}_\mu\rangle = \frac{1}{16\pi^2}\big(\alpha C^2+\beta E+\gamma \Box R\big),$$

or equivalently

$$\langle T^\mu{}_\mu\rangle = \frac{1}{(4\pi)^2}\left(c\, C_{\mu\nu\rho\sigma}^2-a\,E_4+b\,\Box R\right).$$

The geometric invariants are

$$E_4=R_{\mu\nu\rho\sigma}R^{\mu\nu\rho\sigma}-4R_{\mu\nu}R^{\mu\nu}+R^2,$$

$$C_{\mu\nu\rho\sigma}^2=R_{\mu\nu\rho\sigma}R^{\mu\nu\rho\sigma}-2R_{\mu\nu}R^{\mu\nu}+\frac13R^2,$$

and the combination

$\mathcal E_4 \equiv E-\frac{2}{3}\Box R$

is central because of its simple Weyl transformation law (Barvinsky et al., 2023).

The conformally covariant fourth-order scalar operator is the Paneitz operator,

$$\Delta_4 = \Box^2 +2R^{\mu\nu}\nabla_\mu\nabla_\nu -\frac23 R\,\Box +\frac13(\nabla^\mu R)\nabla_\mu,$$

which governs the nonlocal anomaly action and its local conformal-factor representations (Barvinsky et al., 2023). In the same formalism, a canonical nonlocal action is the Riegert–Fradkin–Tseytlin form

$E_4$0

This is the nonlocal functional most directly associated with anomaly-induced gravity in the Nojiri–Odintsov sense (Barvinsky et al., 2023).

A local conformal-factor representation is also available. In the parameterization

$E_4$1

integration of the trace anomaly yields a four-dimensional Liouville or Wess–Zumino action for the conformal mode. In the formulation based on conformal gravity and BRST consistency, the final local action is

$E_4$2

This representation makes explicit the conformal-mode dynamics encoded by the anomaly and the role of the combination $E_4$3 (Kawai et al., 24 Nov 2025).

2. Microscopic field-theoretic mechanism: anomaly poles and renormalization

The field-theoretic underpinning of the anomaly mechanism is exhibited most sharply in the $E_4$4 correlator. In the 13-form-factor $E_4$5-basis,

$E_4$6

with

$E_4$7

The decisive trace relation is

$E_4$8

while only one form factor requires renormalization,

$E_4$9

Taking $\Box R$0 leaves a finite pole term,

$\Box R$1

so the conformal anomaly appears as a massless $\Box R$2 exchange. The paper’s interpretation is that the pole is ultraviolet in origin, because it comes from renormalization, but infrared in appearance, because it survives as a long-range nonlocal kernel (Coriano et al., 2018).

In QED this yields a nonlocal 1PI contribution of the form

$\Box R$3

This is the basic anomaly-pole structure: a traceful gravitational insertion coupled nonlocally through $\Box R$4 to $\Box R$5 (Coriano et al., 2018).

The same conclusion persists in higher stress-tensor correlators. At $\Box R$6 level, renormalization induces “sequential bilinear graviton-scalar mixings on single, double and multiple trace terms, corresponding to $\Box R$7 interactions of the scalar curvature, with intermediate virtual massless exchanges.” At that order a new traceless component also appears after renormalization, so the quartic anomaly sector is not exhausted by the simplest pole terms alone (Corianò et al., 2021).

A complementary exact-CWI analysis reaches the same qualitative result. Exact momentum-space conformal Ward identities and perturbative 1PI amplitudes match at 3-point level, and both predict massless scalar exchanges in $\Box R$8 and $\Box R$9 as the signature of the conformal anomaly (Corianò et al., 2019).

3. Nonlocal action, local dilaton forms, and renormalization-group interpretation

The anomaly-induced action is not unique as a full functional. A useful decomposition is

$\Delta_4$0

where $\Delta_4$1 generates the anomaly and $\Delta_4$2 is Weyl invariant. One may shift

$\Delta_4$3

for any Weyl-invariant functional $\Delta_4$4. This means that the anomaly fixes the Weyl variation, not a unique complete effective action (Barvinsky et al., 2023).

A conformal-orbit construction makes this nonuniqueness explicit. Writing

$\Delta_4$5

the effective action along the orbit yields a Wess–Zumino functional, and different choices of conformal gauge generate a family $\Delta_4$6 of anomaly actions. The classic Riegert–Fradkin–Tseytlin action and the Fradkin–Vilkovisky action are special cases inside this broader family. A central consequence is that the anomaly-induced action alone is not the full effective action; a complementary Weyl-invariant part remains (Barvinsky et al., 2023).

The same issue appears in a flat-space perturbative reconstruction. The anomaly effective action is characterized by bilinear nonlocal mixings

$\Delta_4$7

or, in curved-space language, by structures equivalent to

$\Delta_4$8

supplemented by local Weyl-invariant terms. This is the modern field-theoretic statement of why the anomaly mechanism generates nonlocal gravitational functionals rather than only local higher-curvature corrections (Corianò et al., 2024).

Within an exact renormalization-group framework, the anomaly is the uncancellable vacuum contribution in the Weyl/ERG Ward identity. In $\Delta_4$9, the integrated anomaly

$$\langle T^\mu{}_\mu\rangle = \frac{1}{16\pi^2}\big(\alpha C^2+\beta E+\gamma \Box R\big),$$0

and the Polyakov action

$$\langle T^\mu{}_\mu\rangle = \frac{1}{16\pi^2}\big(\alpha C^2+\beta E+\gamma \Box R\big),$$1

are recovered explicitly. At finite cutoff the Wilsonian action remains quasi-local, while the nonlocal anomaly-induced action survives only in the infrared limit $$\langle T^\mu{}_\mu\rangle = \frac{1}{16\pi^2}\big(\alpha C^2+\beta E+\gamma \Box R\big),$$2. This suggests that the nonlocal kernels used in anomaly-induced gravity are low-energy remnants of integrating out conformal matter in curved space (Rosten, 2018).

4. Cosmological implementation

In cosmological applications the mechanism is inserted into the semiclassical Einstein equations through the quantum stress tensor of conformal matter,

$$\langle T^\mu{}_\mu\rangle = \frac{1}{16\pi^2}\big(\alpha C^2+\beta E+\gamma \Box R\big),$$3

with the anomaly piece written as

$$\langle T^\mu{}_\mu\rangle = \frac{1}{16\pi^2}\big(\alpha C^2+\beta E+\gamma \Box R\big),$$4

For flat FRW one has

$$\langle T^\mu{}_\mu\rangle = \frac{1}{16\pi^2}\big(\alpha C^2+\beta E+\gamma \Box R\big),$$5

so the relevant anomaly terms are the Gauss–Bonnet and $$\langle T^\mu{}_\mu\rangle = \frac{1}{16\pi^2}\big(\alpha C^2+\beta E+\gamma \Box R\big),$$6 sectors (Oikonomou, 12 Mar 2026).

A specific recent realization invokes the Nojiri–Odintsov conformal anomaly mechanism in an analytic slow-roll model that classically ends in a Type II pressure singularity. Near the classical singular time $$\langle T^\mu{}_\mu\rangle = \frac{1}{16\pi^2}\big(\alpha C^2+\beta E+\gamma \Box R\big),$$7, the background behaves as

$$\langle T^\mu{}_\mu\rangle = \frac{1}{16\pi^2}\big(\alpha C^2+\beta E+\gamma \Box R\big),$$8

and the relevant geometric quantities scale as

$$\langle T^\mu{}_\mu\rangle = \frac{1}{16\pi^2}\big(\alpha C^2+\beta E+\gamma \Box R\big),$$9

Because $$\langle T^\mu{}_\mu\rangle = \frac{1}{(4\pi)^2}\left(c\, C_{\mu\nu\rho\sigma}^2-a\,E_4+b\,\Box R\right).$$0 is more singular than $$\langle T^\mu{}_\mu\rangle = \frac{1}{(4\pi)^2}\left(c\, C_{\mu\nu\rho\sigma}^2-a\,E_4+b\,\Box R\right).$$1 in this range, the dominant anomaly contribution is the $$\langle T^\mu{}_\mu\rangle = \frac{1}{(4\pi)^2}\left(c\, C_{\mu\nu\rho\sigma}^2-a\,E_4+b\,\Box R\right).$$2 term; if

$$\langle T^\mu{}_\mu\rangle = \frac{1}{(4\pi)^2}\left(c\, C_{\mu\nu\rho\sigma}^2-a\,E_4+b\,\Box R\right).$$3

then

$$\langle T^\mu{}_\mu\rangle = \frac{1}{(4\pi)^2}\left(c\, C_{\mu\nu\rho\sigma}^2-a\,E_4+b\,\Box R\right).$$4

near the would-be singularity, and the classical branch ceases to satisfy the trace equation. In that implementation, the anomaly therefore erases the classical sudden singularity before the turnaround is reached (Oikonomou, 12 Mar 2026).

The same paper associates the anomaly-dominated regime with intense particle creation and reheating, using the estimate

$$\langle T^\mu{}_\mu\rangle = \frac{1}{(4\pi)^2}\left(c\, C_{\mu\nu\rho\sigma}^2-a\,E_4+b\,\Box R\right).$$5

and argues that radiation production becomes large because $$\langle T^\mu{}_\mu\rangle = \frac{1}{(4\pi)^2}\left(c\, C_{\mu\nu\rho\sigma}^2-a\,E_4+b\,\Box R\right).$$6 while $$\langle T^\mu{}_\mu\rangle = \frac{1}{(4\pi)^2}\left(c\, C_{\mu\nu\rho\sigma}^2-a\,E_4+b\,\Box R\right).$$7 remains finite near the classical Type II singularity. This is a specific use of the anomaly mechanism, not a general theorem about all anomaly-induced cosmologies (Oikonomou, 12 Mar 2026).

A broader early-universe perspective holds that anomaly corrections during inflation scale roughly as $$\langle T^\mu{}_\mu\rangle = \frac{1}{(4\pi)^2}\left(c\, C_{\mu\nu\rho\sigma}^2-a\,E_4+b\,\Box R\right).$$8, that exact de Sitter is conformally flat so $$\langle T^\mu{}_\mu\rangle = \frac{1}{(4\pi)^2}\left(c\, C_{\mu\nu\rho\sigma}^2-a\,E_4+b\,\Box R\right).$$9 on the background, and that the $$E_4=R_{\mu\nu\rho\sigma}R^{\mu\nu\rho\sigma}-4R_{\mu\nu}R^{\mu\nu}+R^2,$$0 sector becomes especially relevant for deviations from exact de Sitter and for tensor perturbations. This suggests that anomaly backreaction is most naturally significant in high-curvature early-universe regimes (Corianò et al., 2024).

5. Alternative formulations, ambiguities, and critiques

The anomaly mechanism is not interpreted uniformly across the literature. Lucat and Prokopec argue that local Weyl symmetry is not genuinely anomalous if the theory includes a Weyl gauge field, naturally identified with the torsion trace, and if regularization is chosen to respect Weyl symmetry. In that framework the fundamental local Ward identity is not

$$E_4=R_{\mu\nu\rho\sigma}R^{\mu\nu\rho\sigma}-4R_{\mu\nu}R^{\mu\nu}+R^2,$$1

but rather

$$E_4=R_{\mu\nu\rho\sigma}R^{\mu\nu\rho\sigma}-4R_{\mu\nu}R^{\mu\nu}+R^2,$$2

The implication is that what is usually called the conformal anomaly may instead reflect omission of the Weyl compensator or a non-Weyl-covariant regularization choice. This is a direct conceptual challenge to the standard interpretation underlying anomaly-induced gravity and cosmology (Lucat et al., 2017).

A different revisionary proposal appears in Conformal Dilaton Gravity. There the Weyl anomaly starts at two loops and is attributed to evanescent operators, but the resulting anomaly is argued to be removable by finite logarithmic counterterms. The general loop-$$E_4=R_{\mu\nu\rho\sigma}R^{\mu\nu\rho\sigma}-4R_{\mu\nu}R^{\mu\nu}+R^2,$$3 finite counterterm is

$$E_4=R_{\mu\nu\rho\sigma}R^{\mu\nu\rho\sigma}-4R_{\mu\nu}R^{\mu\nu}+R^2,$$4

and the full modified action is claimed to satisfy

$$E_4=R_{\mu\nu\rho\sigma}R^{\mu\nu\rho\sigma}-4R_{\mu\nu}R^{\mu\nu}+R^2,$$5

In that setting anomaly-related terms remain important in the infrared because they carry no inverse power of an external mass scale (Álvarez et al., 2015).

Scheme dependence is also deeper than the familiar $$E_4=R_{\mu\nu\rho\sigma}R^{\mu\nu\rho\sigma}-4R_{\mu\nu}R^{\mu\nu}+R^2,$$6 ambiguity. In a vector model formulated to be conformal in $$E_4=R_{\mu\nu\rho\sigma}R^{\mu\nu\rho\sigma}-4R_{\mu\nu}R^{\mu\nu}+R^2,$$7 by introducing an auxiliary scalar compensator,

$$E_4=R_{\mu\nu\rho\sigma}R^{\mu\nu\rho\sigma}-4R_{\mu\nu}R^{\mu\nu}+R^2,$$8

the $$E_4=R_{\mu\nu\rho\sigma}R^{\mu\nu\rho\sigma}-4R_{\mu\nu}R^{\mu\nu}+R^2,$$9 limit leaves a nontrivial scalar remnant in the anomaly and in the anomaly-induced action. The anomaly acquires extra structures such as

$$C_{\mu\nu\rho\sigma}^2=R_{\mu\nu\rho\sigma}R^{\mu\nu\rho\sigma}-2R_{\mu\nu}R^{\mu\nu}+\frac13R^2,$$0

and the localized anomaly-induced action contains effectively three scalar fields rather than the two auxiliary scalars of the standard construction. A plausible implication is that anomaly-induced semiclassical gravity is not completely universal but depends on how the conformal theory is extended away from four dimensions (Oliveira et al., 19 May 2026).

These objections do not eliminate the standard Nojiri–Odintsov mechanism, but they show that its interpretation depends on regularization, on the treatment of compensators or torsion, and on whether the anomaly is regarded as fundamental, removable, or formulation dependent.

6. Extensions: holography, boundaries, and anomaly data

A holographic extension derives the full four-dimensional boundary trace anomaly from five-dimensional scalar–tensor gravities obtained by Kaluza–Klein reduction of the heterotic string effective action. In that setting one finds the standard $$C_{\mu\nu\rho\sigma}^2=R_{\mu\nu\rho\sigma}R^{\mu\nu\rho\sigma}-2R_{\mu\nu}R^{\mu\nu}+\frac13R^2,$$1- and $$C_{\mu\nu\rho\sigma}^2=R_{\mu\nu\rho\sigma}R^{\mu\nu\rho\sigma}-2R_{\mu\nu}R^{\mu\nu}+\frac13R^2,$$2-type curvature coefficients together with scalar-dependent anomaly structures, and a holographic $$C_{\mu\nu\rho\sigma}^2=R_{\mu\nu\rho\sigma}R^{\mu\nu\rho\sigma}-2R_{\mu\nu}R^{\mu\nu}+\frac13R^2,$$3-theorem can be established for the Lovelock–Horndeski branch. This places the anomaly mechanism within a string-theoretic and AdS/CFT setting and continues a line of work explicitly connected to earlier holographic conformal-anomaly studies by Nojiri and Odintsov (Wu et al., 2 May 2025).

Boundaries modify the integrated anomaly in a qualitatively important way. In $$C_{\mu\nu\rho\sigma}^2=R_{\mu\nu\rho\sigma}R^{\mu\nu\rho\sigma}-2R_{\mu\nu}R^{\mu\nu}+\frac13R^2,$$4,

$$C_{\mu\nu\rho\sigma}^2=R_{\mu\nu\rho\sigma}R^{\mu\nu\rho\sigma}-2R_{\mu\nu}R^{\mu\nu}+\frac13R^2,$$5

so boundary conformal invariants built from the extrinsic curvature $$C_{\mu\nu\rho\sigma}^2=R_{\mu\nu\rho\sigma}R^{\mu\nu\rho\sigma}-2R_{\mu\nu}R^{\mu\nu}+\frac13R^2,$$6 become part of the anomaly data. Holographic BCFT calculations suggest that a minimal-surface prescription reproduces the boundary anomaly of $$C_{\mu\nu\rho\sigma}^2=R_{\mu\nu\rho\sigma}R^{\mu\nu\rho\sigma}-2R_{\mu\nu}R^{\mu\nu}+\frac13R^2,$$7 SYM with the $$C_{\mu\nu\rho\sigma}^2=R_{\mu\nu\rho\sigma}R^{\mu\nu\rho\sigma}-2R_{\mu\nu}R^{\mu\nu}+\frac13R^2,$$8-BPS boundary condition more naturally than the alternative tension-parameter prescription (Astaneh et al., 2017).

Finally, explicit computations of anomaly coefficients in cosmologically relevant backgrounds remain important input for anomaly-induced gravity. Adiabatic regularization of a $$C_{\mu\nu\rho\sigma}^2=R_{\mu\nu\rho\sigma}R^{\mu\nu\rho\sigma}-2R_{\mu\nu}R^{\mu\nu}+\frac13R^2,$$9 gauge field in a conformally flat spacetime reproduces the known Maxwell conformal anomaly and yields, in that setting,

$\mathcal E_4 \equiv E-\frac{2}{3}\Box R$0

The regularization-independent part agrees with the standard Maxwell result, while the $\mathcal E_4 \equiv E-\frac{2}{3}\Box R$1 coefficient exhibits the expected gauge and scheme dependence (Chu et al., 2016).

Taken together, these developments indicate that the Nojiri–Odintsov conformal anomaly mechanism is best understood not as a single formula, but as a structured framework. Its stable core is the anomaly-induced coupling of gravitational trace degrees of freedom to nonlocal kernels built from $\mathcal E_4 \equiv E-\frac{2}{3}\Box R$2, $\mathcal E_4 \equiv E-\frac{2}{3}\Box R$3, and $\mathcal E_4 \equiv E-\frac{2}{3}\Box R$4; its unsettled margins concern uniqueness, localization, compensators, regularization dependence, and the physical status of the effective scalar modes signaled by anomaly poles.