SMEFT: Standard Model Effective Field Theory

• SMEFT is a model-independent framework that extends the Standard Model Lagrangian with higher-dimensional, gauge-invariant operators to capture low-energy imprints of heavy new physics.
• It organizes operator contributions by mass-dimension (e.g., dimension-5, -6, and -8) and employs power counting and renormalization group evolution to map UV effects to measurable observables.
• SMEFT underpins global fits across electroweak, Higgs, and flavor sectors, guiding precision studies and high-energy collider searches for beyond Standard Model signals.

The Standard Model Effective Field Theory (SMEFT) is a systematic extension of the Standard Model (SM) that captures the low-energy effects of heavy new physics which is not directly accessible. SMEFT supplements the renormalizable SM Lagrangian with an infinite series of higher-dimensional, gauge-invariant local operators, suppressed by inverse powers of a cutoff scale $\Lambda$. This approach enables a model-independent description of possible deviations from SM predictions, parametrized by Wilson coefficients, and provides a unified language for analyzing precision experimental data and interpreting constraints on possible ultraviolet (UV) completions.

1. Structure of the SMEFT Lagrangian

The SMEFT Lagrangian is organized as a mass-dimension expansion: $$\mathcal{L}_\mathrm{SMEFT} = \mathcal{L}_\mathrm{SM} + \frac{1}{\Lambda}\,C^{(5)}_i \mathcal{O}^{(5)}_i + \frac{1}{\Lambda^2}\,C^{(6)}_i \mathcal{O}^{(6)}_i + \frac{1}{\Lambda^4}\,C^{(8)}_i \mathcal{O}^{(8)}_i + \cdots$$ where $\mathcal{L}_\mathrm{SM}$ denotes the renormalizable SM Lagrangian, $\mathcal{O}^{(d)}_i$ are dimension-$d$ operators constructed from SM fields and covariant derivatives, and $C^{(d)}_i$ are Wilson coefficients encapsulating the effects of new heavy physics. The cutoff scale $\Lambda$ represents the energy scale of new states that have been integrated out.

• Dimension-5: The unique Weinberg operator generates Majorana neutrino masses and violates lepton number by two units.
• Dimension-6: The complete, non-redundant basis (“Warsaw basis” [Grzadkowski et al.]) contains 59 baryon-number-conserving operators for three generations, and additional baryon-number-violating structures. These can be organized into classes by field content: purely bosonic (e.g. $X^3$, $H^6$, $H^4 D^2$, $X^2 H^2$), Yukawa-like, dipole, currents ($\psi^2 H^2 D$), and four-fermion operators.
• Dimension-8 and higher: Thousands of operators, increasingly relevant at high energies or where lower-dimensional terms are absent by symmetry.

All SMEFT operators are required to respect the full $SU(3)_c \times SU(2)_L \times U(1)_Y$ gauge symmetry, and—absent explicit sources—conserve baryon and lepton number at dimensions $d<5$ (Isidori et al., 2023).

2. Power Counting, Operator Bases, and Symmetry Structure

Power Counting: Each operator is suppressed by inverse powers of $\Lambda$, with $\mathcal{O}^{(d)}$ scaling as $1/\Lambda^{d-4}$. Naive Dimensional Analysis (NDA) further refines expectations for naturalness and the size of Wilson coefficients, incorporating loop factors and couplings, e.g., $C_i \sim O(1)$ indicates strongly coupled UV dynamics (Brivio et al., 2017).

Operator Bases: Several bases are used:

• Warsaw basis: Designed for generality and completeness, preferred for global fits and consistent RGE evolution [Grzadkowski et al.].
• EGGM, SILH, HISZ: Tailored to specific physical processes (Higgs, gauge couplings). Field redefinitions relate all bases; physical observables are basis-independent (Henning et al., 2014).

Symmetry Constraints: Gauge invariance and background field Ward identities enforce linear relations among operators, so only certain combinations impact independent physical quantities (e.g., only two combinations of $C_{HW}$, $C_{HB}$, $C_{HWB}$ affect $W$/$Z$ kinetic mixing) (Corbett et al., 2019). Accidental global symmetries (such as flavor $U(3)^5$, $U(2)^5$, or custodial symmetry) may restrict allowed operator flavor structures, suppressing flavor-changing and custodial symmetry-violating amplitudes. Minimal Flavor Violation (MFV) further constrains Wilson coefficients to be proportional to appropriate Yukawa structures, suppressing FCNCs (Isidori et al., 2023).

3. Matching, Renormalization, and Running

Matching: UV models with heavy states are matched onto SMEFT at scale $\Lambda$ by integrating out heavy fields, producing effective operators:

• Tree-level matching: For a linear coupling (e.g., scalar $S$ with $\Lambda_1 S (H^\dagger H)$), integrating out $S$ yields $C_{H\Box} \sim \Lambda_1^2/M_S^4$ (Brivio et al., 2017, Henning et al., 2014).
• One-loop matching: Functional techniques (e.g., Covariant Derivative Expansion) yield general analytic results for any heavy field content (Henning et al., 2014, Chiang et al., 2015, Huo, 2015).

Renormalization Group Evolution (RGE): Below $\Lambda$, Wilson coefficients run with scale according to one-loop anomalous dimension matrices:

$$\mu \frac{d}{d\mu} C_i = \frac{1}{16\pi^2} \gamma_{ij} C_j$$

The full $\gamma_{ij}$ has been computed for the Warsaw basis [Jenkins, Manohar, Trott; (Brivio et al., 2017, Celis et al., 2017)], encompassing operator mixing and threshold corrections. For a consistent global analysis, Wilson coefficients must be matched at the UV scale and evolved to the relevant low scale, properly accounting for operator mixing and induced contributions (especially to strongly constrained directions) (Dawson et al., 2020).

NLO SMEFT: One-loop corrections (including NLO renormalization and EFT matching) are essential for percent-level predictions, reduce scale uncertainty, and ensure field and parameter counterterms absorb all UV divergences up to chosen order (Passarino, 2016).

4. Phenomenological Applications and Global Fits

SMEFT provides a unified language for analyzing experimental data from multiple sectors:

• Electroweak Precision Observables (EWPO): LEP-I Z-pole pseudo-observables ($\Gamma_Z$, $R_\ell^0$, $A_{FB}$, etc.) and W mass measurements stringently constrain combinations of $O_{H\ell}$, $O_{He}$, $O_{Hq}$, $O_{Hu}$, $O_{Hd}$, $O_{HWB}$, $O_{HD}$, and four-fermion operators, typically at the per-mille level (Brivio et al., 2017, Berthier et al., 2015).
• Higgs & Diboson: Higgs couplings, decay rates, and production cross sections constrain $O_{HW}$, $O_{HB}$, $O_{HWB}$, $O_{HG}$, $O_{Hq}^{(1,3)}$, see (Madigan, 2022, Ellis, 2021).
• LHC Run 2 & Forward Physics: High-$p_T$ distributions and multi-boson final states probe energy-enhanced operators, motivating refined power counting in $E/\Lambda$ (Assi et al., 14 Apr 2025).
• Flavor Observables (LEFT): SMEFT is matched at $\mu\sim m_W$ onto Low-Energy Effective Field Theory (LEFT), enabling a single global fit of flavor-conserving and flavor-violating observables (Celis et al., 2017).

Global Fit Methodology:

$$\chi^2(C) = (\vec{O}_\textrm{exp} - \vec{O}_\mathrm{th}(C))^T V^{-1} (\vec{O}_\textrm{exp} - \vec{O}_\mathrm{th}(C))$$

where $V$ is the covariance matrix including theoretical and experimental uncertainties. Precision data are incorporated including cross-correlations and systematic errors; statistical profiling or marginalization over all $C_i$ accounts for operator correlations (Madigan, 2022, Berthier et al., 2015).

Representative Fit Results:

• Marginalized 95% CL bounds on $C_{tG}$, $C_{HG}$, and several top or Higgs operators at the $O(0.05)$–$O(0.2)\ \textrm{TeV}^{-2}$ level, implying $\Lambda \gtrsim 3\,\textrm{TeV}$ (for $C\sim1$), but some four-fermion and flavor sectors remain weakly constrained ($\Lambda \gtrsim 1\,\textrm{TeV}$) (Madigan, 2022, Ellis, 2021).
• Theory errors from neglected dimension-8 operators, loop effects, and matching ambiguities are essential and can expand allowed regions by factors of 2–3 at percent-level precision (Berthier et al., 2015, Dawson et al., 2020).

5. Energy-Enhanced Expansions and High-Energy Phenomenology

At future colliders (e.g., HL-LHC), typical kinematic scales $E$ may approach the new-physics scale $\Lambda$. SMEFT contributions must then be organized according to both $v/\Lambda$ and $E/\Lambda$ expansions. Operators whose matrix elements grow with additional powers of $E$ (“energy-enhanced operators”) dominate the high-energy tails of distributions (Assi et al., 14 Apr 2025). The most relevant subset for high-multiplicity or high-$p_T$ processes is sharply reduced by this scaling.

• Dual expansion: Amplitudes scale as $v^p E^q/\Lambda^{d-4}$, with careful kinematic counting based on field/derivative content. For a fixed process, operators with the highest $E$-weight content are prioritized for interpretation and fitting.
• Implications: Streamlines the parameter space for global fits; clarifies theory validity region ($E<\Lambda$); guides experimental strategies for maximal BSM sensitivity in energy tails.

6. On-Shell and Geometric Approaches

On-Shell S-Matrix Construction: The full SMEFT operator basis is equivalently specified as a basis of unfactorizable local on-shell amplitudes (“amplitude basis”). This construction ensures:

• Completeness with respect to little-group and gauge invariants;
• Independence from EOM or integration-by-parts redundancies;
• Basis-independent characterization of operator effects (Huber et al., 2021, Ma et al., 2019).

Geometric SMEFT (“geoSMEFT”): Reformulation in geometric terms defines field-space metrics $h_{IJ}(\phi)$, $g_{AB}(\phi)$, encoding SMEFT corrections as perturbations of kinetic tensors. This approach:

• Automates field-redefinition invariance and operator reduction;
• Maps amplitude corrections to geometric invariants (e.g., the field-space Riemann tensor at $O(1/\Lambda^4)$);
• Massively reduces combinatorial redundancy in high-dimension operator bases;
• Directly matches to physical observables at any expansion order (Trott, 2022).

7. Applications, Limitations, and Extensions

Benchmark UV Scenarios: SMEFT matching systematically maps tree-level and loop-level UV constructions (e.g., scalar singlets/triplets/doublets, vectorlike fermions, vector bosons, supersymmetric stops) onto boundaries for Wilson coefficients, interpreted in terms of new-physics mass and coupling scales (Chiang et al., 2015, Huo, 2015, Madigan, 2022).

Limitations: The EFT expansion is valid only when $E \ll \Lambda$ and corrections from neglected higher-dimensional operators are subdominant (i.e., terms beyond $v^2/\Lambda^2$, $E^2/\Lambda^2$ are controllable). In certain scenarios (notably strong first-order electroweak phase transition for baryogenesis), operator hierarchies may collapse and higher-dimension effects (dimension-8 or beyond) become as large as leading corrections, invalidating the EFT expansion (Vries et al., 2017). In such contexts, explicit UV models or extended EFT resummations are necessary.

Baryon and Lepton Number Violation: SMEFT includes $\Delta L=2$ and $\Delta B=1$ operators at $d=5$ and $d=6$, respectively. Experimental bounds on proton decay and lepton-number violation push the new-physics scale for these effects to $O(10^{16})$ GeV and $O(10^{14})$ GeV, motivating the neglect of these operators in most collider and low-energy fits (Isidori et al., 2023).

Positivity and Unitarity Constraints: Fundamental principles of analyticity and unitarity constrain combinations and signs of Wilson coefficients. Forward elastic amplitude positivity yields model-independent inequalities for subsets of dimension-8 coefficients (Isidori et al., 2023). EFT unitarity is maintained as long as the expansion parameter $s/\Lambda^2\ll 1$ (Degrande et al., 2012).

For detailed presentations, operator tables, fit results, and further developments, see:

• (Isidori et al., 2023) The Standard Model effective field theory at work
• (Brivio et al., 2017) The Standard Model as an Effective Field Theory
• (Madigan, 2022) Top, Higgs, Diboson and Electroweak Fit to the Standard Model Effective Field Theory
• (Huber et al., 2021) Standard Model EFTs via On-Shell Methods
• (Assi et al., 14 Apr 2025) Energy-Enhanced Expansion of the Standard Model Effective Field Theory
• (Trott, 2022) The geometric SMEFT
• (Henning et al., 2014) How to use the Standard Model effective field theory
• (Berthier et al., 2015) Consistent constraints on the Standard Model Effective Field Theory
• (Passarino, 2016) NLO Standard model effective field theory for Higgs and EW precision data
• (Dawson et al., 2020) Putting SMEFT Fits to Work
• (Celis et al., 2017) DsixTools: The Standard Model Effective Field Theory Toolkit
• (Chiang et al., 2015, Huo, 2015) UV matching for heavy scalar/vectorlike fermion models

---

Summary Table: Warsaw Basis Operator Classes (Dimension-6, flavor/generation generalization suppressed)

Class	Representative Operator(s)	Physical Effects
Pure Higgs	$(H^\dagger H)^3$	Higgs self-coupling, vacuum expectation value
Higgs-deriv.	$(H^\dagger H)\Box(H^\dagger H)$, $|H^\dagger D_\mu H|^2$	Higgs kinetic, EW precision, $T$ parameter
Gauge-Higgs	$(H^\dagger H) X_{\mu\nu} X^{\mu\nu}$	$h\to\gamma\gamma$, $gg\to h$, TGCs
Triple-gauge	$f^{ABC} G^A_{\mu\nu} G^{B\nu\rho} G^{C\rho\mu}$	QCD, EW triple vertices
Yukawa-shift	$(H^\dagger H)(\bar\psi \psi H)$	Higgs–fermion couplings
Dipole	$(\bar q \sigma^{\mu\nu} t) \tilde H G_{\mu\nu}$	Fermion dipoles, $g-2$, FCNC
Current	$$i(H^\dagger \overleftrightarrow{D}_\mu H)(\bar q \gamma^\mu q)$$	$Z$, $W$ couplings
Four-fermion	$$(\bar\psi \gamma_\mu \psi)(\bar\psi \gamma^\mu \psi)$$	Contact interactions, semi-leptonic/diquark