Higher-Spin and Vector-Model Dualities

The higher-spin/vector-model proposal relates the singlet sector of a large-$N$ three-dimensional vector CFT to an interacting theory in AdS$_4$ containing a scalar and an infinite tower of massless gauge fields. Its unusually dense light spectrum is the point, not a defect to be hidden: this is a controlled non-Einstein holographic regime. Matching spins alone is insufficient; scalar quantization, correlator normalization, parity phase, global singlet projection, and the order of the $1/N$ expansion are part of the dictionary.

Required background. Large-N CFT Data and Vector Models supplies current normalization and factorization; Higher-Spin Gaps and Einstein-Regime Obstructions explains why an infinite conserved-current tower forbids a low-spin Einstein EFT.

Helpful background. Vector Models, Auxiliary Fields, and Large-N Saddles develops the boundary expansion; Evidence Programs for Holographic Duality supplies the distinction between a proposed dictionary and an equivalence theorem.

The minimal type-A dictionary

For $N$ free real scalars in $d=3$, the $O(N)$-singlet primaries include

$$J_0=\phi^i\phi^i, \qquad J_{\mu_1\cdots\mu_s} =\phi^i\partial_{(\mu_1}\cdots\partial_{\mu_s)}\phi^i-\text{traces}-\text{descendants},$$

with even $s=2,4,\ldots$, dimensions $\Delta_0=1$ and $\Delta_s=s+1$, and $\partial^{\mu_1}J_{\mu_1\cdots\mu_s}=0$. The proposed minimal type-A bulk theory contains one parity-even scalar with $m^2L^2=-2$ and one massless gauge field for every even spin. The AdS mass–dimension relations give

$$m_0^2L^2=\Delta(\Delta-3)=-2, \qquad m_s^2L^2=(\Delta+s-2)(\Delta-s-1)=0$$

for $\Delta=1$ or $2$ in the scalar case and $\Delta=s+1$ for a conserved current. A $U(N)$ complex-vector singlet sector instead carries all integer spins. Klebanov and Polyakov proposed the critical $O(N)$ model as the alternate scalar boundary condition of the same higher-spin system Klebanov and Polyakov 2002, pp. 213–219.

Normalize boundary currents so that $\langle J_sJ_s\rangle=C_s(N)\mathcal I_s/x^{2\Delta_s}$ with $C_s\sim N$. Canonically normalized bulk fields then have cubic couplings of order $N^{-1/2}$ and loops of order $1/N$, schematically

$$\frac{G_N}{L^2}\sim \frac{1}{N}.$$

The proportionality constant depends on the current and bulk-action normalization and must be fixed by a two-point function. It is not licensed by large-$N$ counting alone.

First application: match the complete linearized tower

Start with the free $O(N)$ model and project to singlets. Conservation fixes each $J_s$ to the short conformal representation $D(s+1,s)$. The corresponding AdS$_4$ Fronsdal field has gauge symmetry

$$\delta\varphi_{\mu_1\cdots\mu_s} =\nabla_{(\mu_1}\xi_{\mu_2\cdots\mu_s)}$$

and the same $SO(3,2)$ representation. $J_2$ maps to the graviton, but it is not isolated: $J_4,J_6,\ldots$ are equally light. The scalar $J_0$ maps to the $m^2L^2=-2$ bulk scalar in alternate quantization, $\Delta_-=1$. Replacing the free fixed point by the critical $O(N)$ singlet sector changes the scalar to standard quantization, $\Delta_+=2$, while the higher-spin currents remain conserved only at leading $N=\infty$.

This spectral match is sharpened by correlators. Tree-level higher-spin calculations reproduce normalized free-vector three-point structures after one overall coupling is fixed Giombi and Yin 2010, §§ 4–6. In Chern–Simons vector models, weakly broken higher-spin Ward identities restrict planar correlators to parity-even boson, parity-even fermion, and parity-odd structures with coupling-dependent coefficients Maldacena and Zhiboedov 2013, §§ 2–5. Those results support a detailed planar dictionary; they do not provide a nonperturbative Hilbert-space equivalence or a conventional local bulk action.

Boundary conditions and parity are physical data

For $m^2L^2=-2$, the near-boundary field is

$$\varphi(z,x)=z\,\alpha(x)+z^2\,\beta(x)+\cdots.$$

Choosing which coefficient is the source exchanges $Δ=1$ and $Δ=2$. Mixed conditions implement multi-trace deformations. Independently, parity-even type A, parity-odd type B, and parity-violating phases select different three-point tensors. Global $O(N)$ versus $U(N)$ data determine the allowed spin tower and singlet projection. These are not decorations on one spectrum.

Adversarial control: hold the spectrum fixed

Keep the scalar mass and higher-spin tower fixed but exchange $α$ and $β$, or turn on a parity phase. The scalar dimension, three-point structures, and RG interpretation change even though every bulk spin is unchanged. Next compare $O(N)$ and $U(N)$ singlets: their towers differ. Any purported dictionary inferred only from the list of masses fails these controls.

The regime is $N\gg1$ with correlators organized in $1/N$ and no parametrically large higher-spin gap. There is no justified truncation to Einstein gravity, no independent small-$\alpha'$ expansion supplied by the Vasiliev description, and no Kaluza–Klein decoupling statement without a top-down embedding. The strongest conclusion is a highly constrained perturbative dictionary with substantial spectrum and correlator support. Nonperturbative completeness, bulk locality, and finite-$N$ equivalence remain separate questions handed to the interaction, correlator, and status analyses.

The chapter overview compares these assumptions and limitations.

For the chapter-wide comparison of assumptions, counterevidence, falsifiers, and claim ceilings, see the claim-domain table.

References

• Giombi, S., and Yin, X. (2010). “Higher Spin Gauge Theory and Holography: The Three-Point Functions.” Journal of High Energy Physics 2010(9), 115. DOI.
• Klebanov, I. R., and Polyakov, A. M. (2002). “AdS Dual of the Critical O(N) Vector Model.” Physics Letters B 550, 213–219. DOI.
• Maldacena, J., and Zhiboedov, A. (2013). “Constraining Conformal Field Theories with a Slightly Broken Higher Spin Symmetry.” Classical and Quantum Gravity 30, 104003. DOI.