Spinning dispersive CFT sum rules and bulk scattering

Cyuan-Han Chang,Yakov Landau,David Simmons-Duffin
2024-05-30
Abstract:We use commutativity of null-integrated operators on the same null plane to construct dispersive CFT sum rules for spinning operators. The contribution of heavy blocks to these sum rules is dominated by a saddle configuration that we call the "scattering crystal." Correlators in this configuration have a natural flat-space interpretation, which allows us to build a dictionary between dispersive CFT sum rules for stress-tensors and flat-space dispersion relations for gravitons. This dictionary is a crucial step for establishing the HPPS conjecture for stress tensor correlators.
High Energy Physics - Theory
What problem does this paper attempt to address?
### Problems the paper attempts to solve This paper aims to solve the following problems by constructing the dispersion conformal field theory (CFT) sum rules of spin operators: 1. **Construction of dispersion CFT sum rules**: - Use the commutation relations of light - like integral operators on the same light - like plane to construct the dispersion CFT sum rules of spin operators. - These sum rules are mainly dominated by a saddle - point configuration in the heavy - block contributions, which the author calls "scattering crystal". 2. **Interpretation of scattering crystal**: - The correlation functions in the scattering crystal have a natural flat - space interpretation, which enables the author to establish a dictionary between the dispersion CFT sum rules and the flat - space dispersion relations. - In particular, a dictionary is established between the dispersion CFT sum rules for the stress - tensor and the flat - space dispersion relations for gravitons. 3. **Verification of the HPPS conjecture**: - Through the above dictionary, the author provides a crucial step for verifying the HPPS conjecture (i.e., large - N CFT should have a local bulk EFT dual when there is a large gap in the single - trace spectrum). - Specifically, the author considers the weakly - coupled EFT in AdS space, which contains massless gravitons and scalar fields with mass \( m_\phi = O\left(\frac{1}{R_{\text{AdS}}}\right) \), and derives the bounds of the higher - derivative interactions \( g_n \). ### Main tools and techniques - **Dispersion CFT sum rules**: These sum rules have double zeros at the positions of the double - twist operators, allowing to separate out the light double - trace contributions and represent them as the sum of heavy conformal blocks with positive coefficients. - **Saddle - point analysis**: In the flat - space limit, the integrals of the dispersion CFT sum rules are completely localized to a saddle - point configuration, and by evaluating the conformally invariant structure at the saddle - point, a concise formula can be obtained. - **Flat - space interpretation**: The inserted conformal blocks become partial waves in flat - space, and the subtracted super - convergent sum rules become the sum rules for the flat - space "impact" amplitudes. ### Structural overview - **Section 2**: Review the super - convergent sum rules and their applications in the scalar case. - **Section 3**: Derive the flat - space limit formulas for scalar and spin operators through space - time saddle - point analysis. - **Section 4**: Explain how to match these formulas to the flat - space sum rules and obtain a dictionary between the spin CFT sum rules and the flat - space sum rules. - **Section 5**: Conclusions and prospects for future work. ### Key formulas - **Dispersion CFT sum rules**: \[ \Psi_{k,\nu}[G] \equiv A_{k,\nu} \int \frac{d^dx_2 d^dx_4 d^dx_0 D^{d - 2}z_0}{\text{vol}(\widetilde{\text{SO}}(d,2))} \langle 0 | \phi_4(x_4^+) L[O](x_0, z_0) \phi_2(x_2) | 0 \rangle^{-1} \] \[ \times \int D^{d - 2}z_1 D^{d - 2}z_3 \langle \tilde{P}_{\delta_1}(z_1) \tilde{P}_{\delta_3}(z_3) P_\delta(z_0) \rangle \langle \Omega | \phi_4(x_4^+) [L[\phi_3](x_0, z_3), L[\phi_1](x_0, z_1)] \phi_2(x_2) | \Omega \rangle \] - **Flat - space limit**: \[ \Psi_{k,\nu;f_k}[\Delta, J] = \frac{2m^2 - \nu^2}{m^2 - \nu^2} P_J\left(1 - \frac{2\nu^2}{m^2}\right) \left(\frac{m^2 (m^2 - \nu^2)}{(m^2 (m^2 -