Dirac Seesaw Mechanism in Neutrino Models

• The Dirac seesaw mechanism is a neutrino mass generation approach where heavy states or small induced vevs suppress the Dirac neutrino mass.
• Models employ diverse variants—type-I-like, type-II-like, and inverse seesaw setups—that use symmetry assignments to forbid Majorana mass terms.
• These frameworks integrate approaches from flavor symmetry, warped geometry, and extended scalar sectors to yield testable phenomenological signatures.

Searching arXiv for relevant papers on Dirac seesaw mechanisms and closely related constructions. arXiv search query: "Dirac seesaw mechanism neutrino masses type-II Dirac inverse seesaw mini-seesaw" Dirac seesaw mechanisms are classes of neutrino-mass constructions in which the smallness of neutrino masses is obtained through a seesaw suppression while the light neutrinos remain Dirac fermions rather than Majorana fermions. In these frameworks, the suppression may arise from heavy intermediary fermions, induced small vacuum expectation values, warped localization effects, radiative generation of Dirac submatrices, or inverse-seesaw-like chains with small symmetry-breaking parameters. A defining structural feature is that lepton number is either exactly conserved or broken in a way that does not induce light Majorana masses, often through residual discrete symmetries or charge assignments that forbid the Weinberg operator and Majorana bilinears. The literature contains several non-equivalent realizations, including warped “mini-seesaw” constructions (McDonald, 2010), type-II Dirac implementations with extended scalar sectors (Bonilla et al., 2017, Berbig, 2022, Oliveira et al., 3 Feb 2025), type-I-like Dirac constructions with extra right-handed states (Matute, 2015, Goswami et al., 30 Jan 2025), inverse-seesaw analogues (Chuliá et al., 2020), and radiative double-seesaw scenarios in left-right models (C. et al., 2024).

1. Conceptual definition and distinguishing criteria

The term “Dirac seesaw” refers to a suppression mechanism in which the effective light-neutrino mass arises from integrating out heavy states or from a parametrically small induced vev, but the resulting light eigenstates are Dirac rather than Majorana. In contrast with the conventional type-I Majorana seesaw, the low-energy spectrum does not contain light Majorana masses and therefore does not predict neutrinoless double-beta decay as a generic consequence (Matute, 2015, Oliveira et al., 3 Feb 2025).

A minimal algebraic template appears in Dirac inverse-seesaw models, where the chiral basis $(\nu_L, N_L)$ versus $(\nu_R, N_R)$ yields

$$\mathcal{M}= \begin{pmatrix} 0 & m_D\ \mu & M_S \end{pmatrix}, \qquad m_\nu \simeq \frac{m_D\,\mu}{M_S},$$

with $\mu \ll m_D \ll M_S$ (Chuliá et al., 2020). Here the smallness of $m_\nu$ is controlled by a small symmetry-breaking parameter $\mu$, so the seesaw suppression is linear in the small breaking term rather than quadratic in a heavy Majorana scale.

A different but equally characteristic pattern appears in type-I-like Dirac constructions with extra right-handed neutrinos. In the model with one “ordinary” and one “adulterant” right-handed neutrino per family, the full mass matrix in the basis $\Psi^T=(\nu_L,V_R^c,U_R^{\prime\,c})$ is

$$\begin{pmatrix} 0 & m_D & m_D'\ m_D & M_R & \mu'\ m_D' & \mu' & M_R' \end{pmatrix},$$

and after integrating out the heavy $V_R$ state one obtains an almost Dirac light pair with

$m_{\rm light}\simeq m_D' - \frac{m_D\,\mu'}{M_R},$

while the heavy eigenstate sits at $(\nu_R, N_R)$0 (Matute, 2015). In the $(\nu_R, N_R)$1 limit the surviving light mass is effectively Dirac: $(\nu_R, N_R)$2

Across the model classes, the operative distinction from Majorana seesaws is structural rather than merely numerical. The suppressions may look “type-I-like,” “type-II-like,” “inverse-seesaw-like,” or “double-seesaw-like,” but the symmetries are arranged so that the light neutrino mass term is of Dirac type. This suggests that “Dirac seesaw” is best viewed as a family of mechanisms rather than a single canonical construction.

2. Canonical mass structures and symmetry protection

A recurring requirement is the absence of allowed Majorana mass terms for the light neutrino sector. Different papers realize this in different ways.

In the flavour-symmetric type-II Dirac model, the symmetry group is $(\nu_R, N_R)$3, where a residual $(\nu_R, N_R)$4 forbids all Majorana mass terms and ensures that neutrinos remain Dirac (Bonilla et al., 2017). The only renormalisable neutrino-mass operator is then the Dirac Yukawa coupling to a new scalar doublet $(\nu_R, N_R)$5, whose vev is itself seesaw-suppressed.

In the mirror-sector type-II Dirac construction, the gauge group $(\nu_R, N_R)$6 contains no gauge-singlet right-handed neutrinos; instead, the mirror neutrino in $(\nu_R, N_R)$7 plays that role (Berbig, 2022). Lepton number remains an exact global symmetry, and the small Dirac mass is induced through a bidoublet scalar $(\nu_R, N_R)$8 with a tiny vev: $(\nu_R, N_R)$9 This is explicitly presented as the Dirac analogue of the type-II seesaw relation (Berbig, 2022).

In the $$\mathcal{M}= \begin{pmatrix} 0 & m_D\ \mu & M_S \end{pmatrix}, \qquad m_\nu \simeq \frac{m_D\,\mu}{M_S},$$0 realization, exact lepton number conservation and a discrete $$\mathcal{M}= \begin{pmatrix} 0 & m_D\ \mu & M_S \end{pmatrix}, \qquad m_\nu \simeq \frac{m_D\,\mu}{M_S},$$1 are imposed. The unique soft $$\mathcal{M}= \begin{pmatrix} 0 & m_D\ \mu & M_S \end{pmatrix}, \qquad m_\nu \simeq \frac{m_D\,\mu}{M_S},$$2-breaking term

$$\mathcal{M}= \begin{pmatrix} 0 & m_D\ \mu & M_S \end{pmatrix}, \qquad m_\nu \simeq \frac{m_D\,\mu}{M_S},$$3

preserves total lepton number but induces a tiny sextet vev

$$\mathcal{M}= \begin{pmatrix} 0 & m_D\ \mu & M_S \end{pmatrix}, \qquad m_\nu \simeq \frac{m_D\,\mu}{M_S},$$4

which then gives

$$\mathcal{M}= \begin{pmatrix} 0 & m_D\ \mu & M_S \end{pmatrix}, \qquad m_\nu \simeq \frac{m_D\,\mu}{M_S},$$5

(Oliveira et al., 3 Feb 2025).

In $$\mathcal{M}= \begin{pmatrix} 0 & m_D\ \mu & M_S \end{pmatrix}, \qquad m_\nu \simeq \frac{m_D\,\mu}{M_S},$$6 models with exotic right-handed neutrino charges $$\mathcal{M}= \begin{pmatrix} 0 & m_D\ \mu & M_S \end{pmatrix}, \qquad m_\nu \simeq \frac{m_D\,\mu}{M_S},$$7, anomaly freedom is maintained while singlet scalars with $$\mathcal{M}= \begin{pmatrix} 0 & m_D\ \mu & M_S \end{pmatrix}, \qquad m_\nu \simeq \frac{m_D\,\mu}{M_S},$$8 break the gauge symmetry in a way that forbids Majorana masses for $$\mathcal{M}= \begin{pmatrix} 0 & m_D\ \mu & M_S \end{pmatrix}, \qquad m_\nu \simeq \frac{m_D\,\mu}{M_S},$$9. The effective light Dirac mass then arises from the off-diagonal block structure

$\mu \ll m_D \ll M_S$0

which is the direct Dirac analogue of a type-I seesaw formula (Ma et al., 2014).

The inverse-seesaw family provides a particularly transparent symmetry logic: the small $\mu \ll m_D \ll M_S$1 parameter is the only source of $\mu \ll m_D \ll M_S$2 breaking to $\mu \ll m_D \ll M_S$3, so $\mu \ll m_D \ll M_S$4 restores the symmetry and makes neutrinos strictly massless Diracs. The smallness of $\mu \ll m_D \ll M_S$5 is therefore technically natural in the sense emphasized in that construction (Chuliá et al., 2020).

3. Type-I, inverse, and extended Dirac seesaw realizations

Type-I-like Dirac constructions use heavy fermions to suppress the effective Dirac mass while retaining a Dirac light spectrum. Several distinct realizations exist.

A direct extension of the Standard Model with two right-handed neutrinos per family separates roles between a high-scale “ordinary” right-handed neutrino $\mu \ll m_D \ll M_S$6 and a low-scale “adulterant” right-handed neutrino $\mu \ll m_D \ll M_S$7. The assumed hierarchy

$\mu \ll m_D \ll M_S$8

leads, after block diagonalization, to a heavy decoupled state at $\mu \ll m_D \ll M_S$9 and an almost Dirac light pair of mass $m_\nu$0 (Matute, 2015). Numerically, the model quotes

$m_\nu$1

and

$m_\nu$2

(Matute, 2015). The resulting low-energy theory contains effectively pure Dirac neutrinos, negligible active-sterile mixing of order $m_\nu$3, and no neutrinoless double-beta decay (Matute, 2015).

A related flavour-model realization uses the group

$m_\nu$4

with heavy vector-like singlets $m_\nu$5 and a Dirac-seesaw mass matrix

$m_\nu$6

The paper develops the consequences of permutations in the charged-lepton diagonalization matrix $m_\nu$7, leading to 18 inequivalent $m_\nu$8 choices and six column-permuted neutrino solutions, of which only $m_\nu$9 and $\mu$0 survive the oscillation-data constraints (Goswami et al., 30 Jan 2025). The model predicts a tightly restricted solar angle,

$\mu$1

for both hierarchies, and specific ranges for the masses in the $\mu$2 or $\mu$3 limits (Goswami et al., 30 Jan 2025).

The Dirac inverse-seesaw family abstracts the essential suppression into the product $\mu$4. One notable feature is that large Yukawas $\mu$5 and TeV-scale $\mu$6 remain compatible with $\mu$7 eV if $\mu$8 is in the eV–10 eV range (Chuliá et al., 2020). The same framework generalizes to “double” and “triple” Dirac inverse seesaws: $\mu$9 respectively (Chuliá et al., 2020).

A distinct “dynamical seesaw” realization in an $\Psi^T=(\nu_L,V_R^c,U_R^{\prime\,c})$0 model obtains Dirac masses from a $\Psi^T=(\nu_L,V_R^c,U_R^{\prime\,c})$1 block in the $\Psi^T=(\nu_L,V_R^c,U_R^{\prime\,c})$2 versus $\Psi^T=(\nu_L,V_R^c,U_R^{\prime\,c})$3 basis,

$\Psi^T=(\nu_L,V_R^c,U_R^{\prime\,c})$4

with the light mass controlled by

$\Psi^T=(\nu_L,V_R^c,U_R^{\prime\,c})$5

Its suppression is dynamical because the small combination $\Psi^T=(\nu_L,V_R^c,U_R^{\prime\,c})$6 follows from a softly lepton-number-breaking trilinear $\Psi^T=(\nu_L,V_R^c,U_R^{\prime\,c})$7 in the scalar potential (Valle et al., 2016). The paper emphasizes three simultaneous suppressions: the hierarchy $\Psi^T=(\nu_L,V_R^c,U_R^{\prime\,c})$8, the small soft parameter $\Psi^T=(\nu_L,V_R^c,U_R^{\prime\,c})$9, and a determinant-like Yukawa alignment $$\begin{pmatrix} 0 & m_D & m_D'\ m_D & M_R & \mu'\ m_D' & \mu' & M_R' \end{pmatrix},$$0 (Valle et al., 2016).

4. Type-II Dirac seesaw and induced small vacuum expectation values

Type-II Dirac seesaws are the closest analogues of the conventional Majorana type-II mechanism. Their central object is a scalar whose induced tiny vev directly generates a Dirac mass.

The flavour-symmetric model with $$\begin{pmatrix} 0 & m_D & m_D'\ m_D & M_R & \mu'\ m_D' & \mu' & M_R' \end{pmatrix},$$1 introduces a new scalar doublet $$\begin{pmatrix} 0 & m_D & m_D'\ m_D & M_R & \mu'\ m_D' & \mu' & M_R' \end{pmatrix},$$2 and singlet $$\begin{pmatrix} 0 & m_D & m_D'\ m_D & M_R & \mu'\ m_D' & \mu' & M_R' \end{pmatrix},$$3, with the crucial trilinear scalar-potential term

$$\begin{pmatrix} 0 & m_D & m_D'\ m_D & M_R & \mu'\ m_D' & \mu' & M_R' \end{pmatrix},$$4

After $$\begin{pmatrix} 0 & m_D & m_D'\ m_D & M_R & \mu'\ m_D' & \mu' & M_R' \end{pmatrix},$$5 acquires a vev, the minimization conditions induce

$$\begin{pmatrix} 0 & m_D & m_D'\ m_D & M_R & \mu'\ m_D' & \mu' & M_R' \end{pmatrix},$$6

and the neutrino Dirac mass matrix becomes

$$\begin{pmatrix} 0 & m_D & m_D'\ m_D & M_R & \mu'\ m_D' & \mu' & M_R' \end{pmatrix},$$7

with explicit $$\begin{pmatrix} 0 & m_D & m_D'\ m_D & M_R & \mu'\ m_D' & \mu' & M_R' \end{pmatrix},$$8-dictated texture

$$\begin{pmatrix} 0 & m_D & m_D'\ m_D & M_R & \mu'\ m_D' & \mu' & M_R' \end{pmatrix},$$9

The same flavour structure also yields the “golden” mass relation

$V_R$0

and the oscillation analysis selects only inverted ordering, non-maximal atmospheric mixing, and a correlation between $V_R$1 and the lightest mass $V_R$2 (Bonilla et al., 2017).

A conceptually different type-II Dirac seesaw arises in the mirror-sector portal model. There the bidoublet

$V_R$3

contains one neutral, two singly charged, and one doubly charged scalar. The trilinear term

$V_R$4

induces

$V_R$5

The construction further embeds a generalized parity, “Higgs-parity,” under which

$V_R$6

forcing $V_R$7 and $V_R$8, with the tree-level quark mass matrices satisfying $V_R$9 (Berbig, 2022). Loop corrections generate only $m_{\rm light}\simeq m_D' - \frac{m_D\,\mu'}{M_R},$0, below the bound $m_{\rm light}\simeq m_D' - \frac{m_D\,\mu'}{M_R},$1 (Berbig, 2022). This model therefore uses the type-II Dirac seesaw not only for neutrino masses but also as a portal to a mirror sector and to a solution of the strong CP problem.

The $m_{\rm light}\simeq m_D' - \frac{m_D\,\mu'}{M_R},$2 model implements the type-II Dirac seesaw with a scalar sextet

$m_{\rm light}\simeq m_D' - \frac{m_D\,\mu'}{M_R},$3

where $m_{\rm light}\simeq m_D' - \frac{m_D\,\mu'}{M_R},$4 is the neutral component whose small vev controls the Dirac mass (Oliveira et al., 3 Feb 2025). The mechanism is explicitly described as “exactly analogous to the Majorana type-II result but here generating a Dirac VEV” through

$m_{\rm light}\simeq m_D' - \frac{m_D\,\mu'}{M_R},$5

For $m_{\rm light}\simeq m_D' - \frac{m_D\,\mu'}{M_R},$6 keV, $m_{\rm light}\simeq m_D' - \frac{m_D\,\mu'}{M_R},$7 TeV, and $m_{\rm light}\simeq m_D' - \frac{m_D\,\mu'}{M_R},$8 GeV, the paper states that one obtains $m_{\rm light}\simeq m_D' - \frac{m_D\,\mu'}{M_R},$9 (Oliveira et al., 3 Feb 2025).

Finally, the “double type-II Dirac seesaw” in a $(\nu_R, N_R)$00 framework uses a TeV-scale Higgs doublet $(\nu_R, N_R)$01 with a tiny induced vev. Heavy scalar doublets $(\nu_R, N_R)$02 and $(\nu_R, N_R)$03 generate the mixing

$(\nu_R, N_R)$04

leading to

$(\nu_R, N_R)$05

The numerical example gives

$(\nu_R, N_R)$06

hence

$(\nu_R, N_R)$07

(Gu, 2019).

5. Warped, radiative, and composite-sector variants

Not all Dirac seesaw ideas rely on conventional weakly coupled heavy singlets or triplets. Some use warped geometry, radiative generation, or hidden strong sectors.

In a warped slice of $(\nu_R, N_R)$08, the model of light neutrinos from a mini-seesaw mechanism employs a bulk singlet 5D Dirac fermion $(\nu_R, N_R)$09 with action

$(\nu_R, N_R)$10

The right-chiral zero mode has profile

$(\nu_R, N_R)$11

and for $(\nu_R, N_R)$12 it is localized near the IR brane and strongly suppressed on the UV brane. This yields a tiny UV-induced Dirac mass

$(\nu_R, N_R)$13

together with an IR-localized Majorana mass

$(\nu_R, N_R)$14

The light eigenvalue is

$(\nu_R, N_R)$15

which the paper characterizes as a “mini-seesaw” with $(\nu_R, N_R)$16 (McDonald, 2010). For $(\nu_R, N_R)$17 TeV, $(\nu_R, N_R)$18 TeV, $(\nu_R, N_R)$19, and $(\nu_R, N_R)$20, the representative values are $(\nu_R, N_R)$21 GeV, $(\nu_R, N_R)$22 GeV, $(\nu_R, N_R)$23 GeV, and $(\nu_R, N_R)$24 eV (McDonald, 2010).

The same paper also gives a 4D dual description via AdS/CFT: the right-handed zero mode is predominantly a CFT composite, the UV overlap explains the tiny effective Dirac coupling, and the IR-brane Majorana term corresponds to explicit lepton-number breaking confined to the hidden CFT sector at scale $(\nu_R, N_R)$25 (McDonald, 2010). Although this construction is not a purely Dirac low-energy theory, it is central to the broader Dirac–Majorana seesaw landscape because it shows how suppressed Dirac and Majorana scales can coexist below the weak scale.

Radiative Dirac submatrices appear prominently in the left-right model with a radiative double seesaw. The gauge group

$(\nu_R, N_R)$26

is augmented by a global $(\nu_R, N_R)$27 that breaks to a remnant $(\nu_R, N_R)$28, forbidding tree-level $(\nu_R, N_R)$29–$(\nu_R, N_R)$30 Yukawa couplings. The Dirac mass arises at one loop: $(\nu_R, N_R)$31 with

$(\nu_R, N_R)$32

The full neutrino mass matrix in the basis $(\nu_R, N_R)$33 is

$(\nu_R, N_R)$34

so that in the seesaw limit

$(\nu_R, N_R)$35

(C. et al., 2024). The article emphasizes that this is the first implementation, to its knowledge, of a radiative double seesaw in which the Dirac submatrix is generated at one loop (C. et al., 2024).

A different use of heavy Dirac mediators appears in the TeV-scale five-plet model. There, a vector-like Dirac five-plet

$(\nu_R, N_R)$36

and scalar four-plets $(\nu_R, N_R)$37, $(\nu_R, N_R)$38 generate the dimension-nine operator

$(\nu_R, N_R)$39

After induced vevs

$(\nu_R, N_R)$40

the neutrino mass scales as

$(\nu_R, N_R)$41

This mechanism uses Dirac mediators but produces Majorana neutrino masses, so it does not belong to the strict class of Dirac-neutrino seesaws (Picek et al., 2011). Its relevance is taxonomic: it illustrates that “Dirac seesaw” may refer either to Dirac mediators or to Dirac light neutrinos, and the two notions should not be conflated.

6. Phenomenology, cosmology, and model-dependent signatures

The phenomenology of Dirac seesaw models is highly model-dependent because the suppression can be tied to different sectors: flavour, gauge extensions, warped geometry, hidden strong dynamics, or scalar portals.

In flavour-symmetric type-II Dirac models, the most distinctive outputs are oscillation correlations rather than direct collider states. The $(\nu_R, N_R)$42 model predicts only inverted ordering, prefers the higher octant $(\nu_R, N_R)$43, and yields

$(\nu_R, N_R)$44

for the lightest neutrino mass (Bonilla et al., 2017). It also implies that for $(\nu_R, N_R)$45 eV one must have $(\nu_R, N_R)$46 (Bonilla et al., 2017).

In the $(\nu_R, N_R)$47 type-I Dirac model, charged-lepton flavour violation and dark matter depend on the same structures that control the PMNS matrix. The paper gives

$(\nu_R, N_R)$48

depending on the parameters and the permutation choice in $(\nu_R, N_R)$49 (Goswami et al., 30 Jan 2025). Its dark-matter annihilation cross sections span from $(\nu_R, N_R)$50 down to $(\nu_R, N_R)$51 (Goswami et al., 30 Jan 2025).

The $(\nu_R, N_R)$52 type-II Dirac model links neutrino physics to both lepton-flavour violation and cosmology. For $(\nu_R, N_R)$53, the doubly charged scalar $(\nu_R, N_R)$54 gives

$(\nu_R, N_R)$55

below the present bound $(\nu_R, N_R)$56 (Oliveira et al., 3 Feb 2025). More stringently, the right-handed neutrinos thermalize through $(\nu_R, N_R)$57-mediated interactions, and requiring $(\nu_R, N_R)$58 yields

$(\nu_R, N_R)$59

(Oliveira et al., 3 Feb 2025).

The double type-II Dirac seesaw with $(\nu_R, N_R)$60 adds a stable Dirac fermionic dark matter state

$(\nu_R, N_R)$61

Its relic-density analysis gives

$(\nu_R, N_R)$62

and direct detection implies $(\nu_R, N_R)$63 TeV, with

$(\nu_R, N_R)$64

(Gu, 2019). The same setup supports Dirac leptogenesis through out-of-equilibrium decays of heavy scalar doublets, with

$(\nu_R, N_R)$65

and a representative asymmetry $(\nu_R, N_R)$66 for $(\nu_R, N_R)$67 CP phases (Gu, 2019).

In the mirror type-II Dirac portal, collider-scale signatures are muted because the bidoublet mass scale is typically large, but the scalar spectrum includes singly and doubly charged states analogous to those in Majorana type-II seesaws. Their couplings to SM leptons are, however, restricted by a $(\nu_R, N_R)$68, and no observable shifts in $(\nu_R, N_R)$69, $(\nu_R, N_R)$70, or the $(\nu_R, N_R)$71-parameter arise because the $(\nu_R, N_R)$72 vev respects custodial symmetry (Berbig, 2022). Cosmologically, the paper states that dark radiation from mirror neutrinos is negligible, with freeze-in yields $(\nu_R, N_R)$73, while avoiding overclosure by stable mirror electrons requires

$(\nu_R, N_R)$74

for $(\nu_R, N_R)$75 GeV (Berbig, 2022).

The warped mini-seesaw offers a different phenomenological message: the representative spectrum places the sterile Majorana scale and the effective Dirac scale in the GeV or sub-GeV range rather than at the canonical high seesaw scale (McDonald, 2010). This suggests that warped/composite realizations can decouple neutrino-mass suppression from ultraheavy mediator masses.

7. Relation to Majorana seesaws, common misconceptions, and theoretical scope

The relation between Dirac and Majorana seesaws is subtle, and the literature uses overlapping terminology. Three distinctions are especially important.

First, a seesaw with Dirac mediators is not necessarily a Dirac-neutrino seesaw. The five-plet model uses vector-like Dirac leptons $(\nu_R, N_R)$76, yet the induced operator is Majorana and the resulting light neutrinos are Majorana (Picek et al., 2011). By contrast, the models in which residual symmetries forbid Majorana mass terms produce genuinely Dirac light neutrinos (Bonilla et al., 2017, Chuliá et al., 2020, Oliveira et al., 3 Feb 2025).

Second, the absence of light Majorana masses does not imply the absence of all $(\nu_R, N_R)$77 structures in the ultraviolet. In the “Dirac lepton number violation” scenario, heavy-sector masses satisfy $(\nu_R, N_R)$78, allowing the heavy eigenstate to be a Dirac fermion $(\nu_R, N_R)$79 of mass

$(\nu_R, N_R)$80

while the effective heavy-sector combination

$(\nu_R, N_R)$81

still carries $(\nu_R, N_R)$82 (Dyatlov, 2020). The paper argues that light neutrinos remain Dirac, but heavy-sector interactions can violate lepton number by two units and may still support leptogenesis (Dyatlov, 2020). This is not the standard symmetry-protected Dirac seesaw paradigm, but it illustrates that “Diracness” at low energies need not coincide with exact lepton-number conservation in the full theory.

Third, not every model marketed as a “Dirac seesaw” uses the same suppression logic. Some are type-II-like and controlled by induced vevs (Bonilla et al., 2017, Berbig, 2022, Oliveira et al., 3 Feb 2025, Gu, 2019). Some are type-I-like and controlled by heavy Dirac blocks (Ma et al., 2014, Goswami et al., 30 Jan 2025). Some are inverse-seesaw-like and controlled by small $(\nu_R, N_R)$83-terms (Chuliá et al., 2020). Some are double-seesaw-like, with a radiative Dirac submatrix embedded inside a Majorana right-handed sector (C. et al., 2024). Some combine Dirac and Majorana suppressions in a mini-seesaw framework (McDonald, 2010).

A plausible implication is that “Dirac seesaw mechanism” should be understood as an umbrella category characterized by the coexistence of a seesaw suppression and a symmetry rationale for preserving the Dirac nature of the light neutrinos. Within that umbrella, model builders trade off between symmetry economy, mediator mass scales, flavour predictivity, collider accessibility, and cosmological structure. The resulting landscape spans ultraviolet scales from the GeV regime in warped mini-seesaws (McDonald, 2010) through TeV-scale scalar and gauge sectors (Gu, 2019, Oliveira et al., 3 Feb 2025) up to mirror-sector or grand-unified scales in flavour and parity constructions (Berbig, 2022, Matute, 2015).