Cosmological Cutting Rules from Flat-Space Unitarity via Dressing
Arhum Ansari, Sachin Jain and Deep Mazumdar · 2026
Read Cosmological Cutting Rules from Flat-Space Unitarity via Dressing by Arhum Ansari, Sachin Jain and Deep Mazumdar (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-08917.txt.
Introduction
Arhum Ansari, Sachin Jain, and Deep Mazumdar link rules from flat space physics to rules for the fast growth era of space. In flat space, a rule set says how a process can split into parts. Those split rules are called cutting rules. In curved, fast growth space, the same splits are more hard to track by hand.
The team gives a dress up step that lifts flat cuts to curved space. In that view, each flat cut maps to a jump in a key energy count. They show the map on tree form and one loop cases for a plain field type. A reader in theory work gets a clear path. It makes curved space cut rules less of a black art.
Catalog record
| CREATOR | Arhum Ansari, Sachin Jain and Deep Mazumdar |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 41 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2601-08917
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Cosmological Cutting Rules from Flat-Space Unitarity via Dressing Arhum Ansari, Sachin Jain and Deep Mazumdar Abstract Using cosmological dressing rules, we uplift flat-space unitarity cuts to discontinuity relations for dS/EAdS observables. In this representation, Cutkosky delta functions map directly to “Disc” operations in the exchanged energy variable. This provides a transparent diagram-by-diagram origin of cosmological cutting rules. We illustrate this with explicit examples at tree level and one loop for conformally coupled scalars. 1 Introduction Understanding how fundamental principles such as unitarity and causality manifest across different spacetime backgrounds is a central theme in quantum field theory. In flat-space, unitarity plays a fundamental role in constraining allowed class of QFTs 20 ; 22 . Unitarity relates the discontinuities of scattering amplitudes to sums over on-shell intermediate states. These rules provide a powerful organizational principle for loop amplitudes and underpin modern developments in amplitude-based methods. Extending this intuition to curved spacetimes remains a subtle problem, particularly in de Sitter (dS) space, where conventional notions of asymptotic states and scattering amplitudes are absent, see 18 . In the dS context, physical information is instead encoded in boundary correlation functions, whose analytic structure carries imprints of bulk dynamics. A number of works have shown that unitarity in cosmology can be packaged into discontinuity relations: at the level of the wavefunction this appears as a cosmological analogue of the optical theorem 19 ; 17 ; 3 ; 16 , while at the correlator level one finds cutting rules that relate discontinuities to products of lower-point building blocks 13 ; 11 . At the same time, the correlator story contains an important aspect: the objects that naturally multiply on the right-hand side are not always the correlators themselves, but particular combinations that do not appear as standalone observables. Independently, cosmological dressing rules 7 provide a direct map from flat-space Feynman integrands to dS/EAdS observables by replacing propagators with simple, dressed kernels. This map is precisely the sort of framework in which one might hope to import flat-space unitarity almost mechanically. Through this note, we add to this story and make connection between these two threads explicit: we show how the familiar flat-space statement “unitarity = Cutkosky cuts” becomes, after dressing and analytic continuation, the discontinuity operations that appear in dS/EAdS correlators. Put differently, we explain how the on-shell delta functions arising from Cutkosky rules uplift into the relevant cosmological “Disc” operations in the exchanged energy variable, and how the auxiliary combinations required at correlator level arise naturally in the dressed representation. Our goal is to provide a derivation of the known cosmological cutting rules that is more closely aligned with the reasoning and techniques used in flat-space amplitude analyses. This produces a clean operational recipe: start from the flat-space optical theorem, express the cut in the standard on-shell language, and then dress the result to obtain the corresponding dS/EAdS discontinuity relation. We illustrate this dictionary on representative exchange and loop diagrams and comment on extensions (including spinning exchange). The paper is organised in the following manner. In section 2 , we review the notions of unitarity in flat as well dS space. We describe the statement of optical theorem in flat and describe the discontinuity relations for cosmological correlators. We also review the dressing rules for conformally coupled scalars in dS (EAdS) space as given in 8 . In section 3 , we use split representation and flat space analouge of cutting, to derive the cosmological cutting rules. Finally in section 4 , we describe how flat-space optical theorem, with appropriate analytic continuation, can be dressed to give discontinuity relations in dS/EAdS. We demonstrate this using tree and loop level processes involving conformally coupled scalars. We finally conclude with some future directions in section 5 . 2 Background In this section, we review some of the background needed for the later part of the paper. This includes the notions of unitarity in Minkowski space and dS space, and the dressing rules. Let us discuss these one by one starting from unitarity in fat space. 2.1 Unitarity in flat space Unitarity of flat space scattering amplitude implies \displaystyle SS^{\dagger}=\mathbf{I}\;. (1) Writing S=\mathbf{I}+i\mathcal{T} , this translates to \displaystyle-i(\mathcal{T}-\mathcal{T}^{\dagger})=\mathcal{T}\mathcal{T}^{\dagger}\;. (2) We can also write ( 2 ) in terms of matrix elements as follows 20 , \displaystyle-i[\mathcal{A}(a\rightarrow b)-\mathcal{A}^{*}(a\rightarrow b)]=\sum_{f}d\Pi_{f}\mathcal{A}^{*}(b\rightarrow f)\mathcal{A}(a\rightarrow f), (3) where a and b are asymptotic states and f is all possible intermediate states and \Pi_{f} is the phase space measure. The equation ( 3 ) is a statement of optical theorem . For the case, when a and b are two particle states, we get \displaystyle-i[\mathcal{A}(k_{1}k_{2}\rightarrow p_{1}p_{2})-\mathcal{A}^{*}(p_{1}p_{2}\rightarrow k_{1}k_{2})]=\sum_{n}\Big(\prod_{i=1}^{n}\int\frac{d^{3}q_{i}}{2|\vec{q}_{i}|}\Big) \displaystyle\mathcal{A}^{*}(p_{1}p_{2}\rightarrow q_{i})\mathcal{A}(k_{1}k_{2}\rightarrow q_{i}) \displaystyle\times(2\pi)^{4}\delta^{4}(k_{1}+k_{2}-\{q_{i}\})\;. (4) 2.2 Unitarity in de Sitter space In dS space, unitarity relates higher point wavefunction coefficients to lower point coefficients known as the Cosmological Optical Theorem (COT) 19 ; 17 . The COT gives rise to a set of cutting rules for wavefunction coefficients. These cutting rules states that, the wavefunction coefficient ( \psi ) factorizes into lower point wavefunction coefficients under the operation of certain discontinuities. Consider the case of four point wavefunction coefficient. A action of discontinuity operation for s-channel results in the following factorization \displaystyle i\textrm{Disc}_{p_{s}}\Big(i\psi_{4}(k_{1},k_{2},k_{3},k_{4})\Big)=\int_{qq^{\prime}}\Big(i\textrm{Disc}_{q}{i\psi_{3}(k_{1},k_{2},q)}\Big)P_{qq^{\prime}}\Big(i\textrm{Disc}_{q^{\prime}}{i\psi_{3}(q^{\prime},k_{3},k_{4})}\Big), (5) where the “Disc” operation is defined as \displaystyle\textrm{Disc}_{k_{i}}f(k_{1},\cdots,k_{i},\cdots,k_{n},\{p\},\{\mathbf{k}\}) \displaystyle\equiv f(k_{1},\cdots,k_{i},\cdots,k_{n},\{p\},\{\mathbf{k}\})-f^{*}(-k_{1},\cdots,k_{i},\cdots,-k_{n},\{p\},\{-\mathbf{k}\}), (6) and P_{qq^{\prime}} is the power spectrum. Applying discontinuity rules to in-in correlator 13 using Schwinger-Keldysh formalism for polynomial interaction one obtains, \displaystyle\mathrm{Disc}_{p}\mathcal{B}^{(2)}(\{k_{L},k_{R}\};p)=\frac{1}{2P_{p}(\eta_{0})}\Big( \displaystyle\mathrm{Disc}_{p}\mathcal{B}^{(1)}(\{k_{L}\},p)\times\mathrm{Disc}_{p}{\mathcal{B}}^{(1)}(\{k_{R}\},p) \displaystyle-\widetilde{\mathrm{Disc}}_{p}\tilde{\mathcal{B}}^{(1)}(\{k_{L}\},p)\times\widetilde{\mathrm{Disc}}_{p}\tilde{\mathcal{B}}^{(1)}(\{k_{R}\},p)\Big), (7) \mathcal{B}^{(2)} is the two-site scalar correlator and \mathcal{B}^{(1)} and \tilde{\mathcal{B}^{(1)}} are defined as \displaystyle\mathcal{B}^{(1)}(k_{1},\cdots,k_{n})=\frac{2\mathbb{R}e\;\psi_{n}(k_{1},\cdots,k_{n})}{\prod_{i=1}^{n}2\mathbb{R}e\;\psi_{2}(k_{i})}, \displaystyle\tilde{\mathcal{B}}^{(1)}(k_{1},\cdots,k_{n})=\frac{2i\mathbb{I}m\;\psi_{n}(k_{1},\cdots,k_{n})}{\prod_{i=1}^{n}2\mathbb{R}e\;\psi_{2}(k_{i})}. (8) Note that there are two types of discontinuity defined in ( 2.2 ) \displaystyle\textrm{Disc}_{p}\mathcal{B}^{(1)}(\{k_{L}\},p) \displaystyle=\mathcal{B}^{(1)}(\{k_{L}\},p)-\mathcal{B}^{(1)}(\{k_{L}\},-p), \displaystyle\widetilde{\textrm{Disc}}_{p}\mathcal{B}^{(1)}(\{k_{L}\},p) \displaystyle=\mathcal{B}^{(1)}(\{k_{L}\},p)+\mathcal{B}^{(1)}(\{k_{L}\},-p). (9) The notation used in equation ( 2.2 ) is given in more detail in equation (4.21) and figure 2 of 13 . With this prelude to cutting rules in cosmology, for both the wavefunction coefficients ( 5 ) and correlators ( 2.2 ), we will now try delve into the origins of such rules for the latter and see if we can learn something from flat space cutting rules 12 . To that end, let us try to understand how the cosmological correlators can be written by ‘dressing’ the flat space amplitudes. 2.3 Dressing rules Recently, there has been significant progress in evaluating cosmological correlators using EAdS formalism 23 ; 10 ; 7 ; 8 ; 1 ; 9 ; 15 . It was illustrated in 8 , that there exists a remarkable connection between the EAdS Witten diagram, and a flat-space Feynman diagram, when the latter is dressed with a theory dependent factor 8 . We shall briefly review this point in this section. Tree level scalar correlators Let us start with the cubic theory of conformally coupled scalar in the EAdS shadow action. \mathbf{\phi}^{3} theory Consider the s-channel exchange diagram shown in Figure 1 , for which there can be two possible exchanges: (+) or ( - ), from \phi_{-}^{2}\phi_{+} and \phi_{-}^{3} vertices respectively 8 . p,\vec{s} \vec{k}_{1} \vec{k}_{2} \vec{k}_{3} \vec{k}_{4} (a) (+) exchange p,\vec{s} \vec{k}_{1} \vec{k}_{2} \vec{k}_{3} \vec{k}_{4} (b) (-) exchange Figure 1: Witten diagrams representing 4-point correlation functions. (a) Exchange via solid (+) propagator. (b) Exchange via a dashed ( - ) propagator. The Witten diagram calculation corresponding to the (+) exchange is as follows \displaystyle\left\langle\phi_{1}(k_{1})\phi_{2}(k_{2})\phi_{3}(k_{3})\phi_{4}(k_{4})\right\rangle_{s}^{+}= \displaystyle g^{2}\int_{0}^{\infty}\frac{dz_{1}}{z_{1}^{4}}\frac{dz_{2}}{z_{2}^{4}}(z_{1}^{2}e^{-k_{12}z_{1}})(z_{2}^{2}e^{-k_{34}z_{2}})\int_{-\infty}^{\infty}\frac{pdp}{p^{2}+s^{2}}(z_{1}z_{2})^{\frac{3}{2}}J_{\frac{1}{2}}(pz_{1})J_{\frac{1}{2}}(pz_{2}) \displaystyle=