On the Inverse Problem in Effective Field Theory

This paper introduces a new class of nonlinear analytic dispersion relations that enable the direct extraction of the tree-level spectrum of heavy particles from the Wilson coefficients of an effective field theory, providing an exact solution for a finite number of resonances and an approximate one otherwise.

The Gist

Imagine you are a detective trying to solve a crime, but you only have the footprints left behind at the scene, not the criminal themselves.

In the world of particle physics, this is exactly the challenge scientists face. They have a theory called Effective Field Theory (EFT), which describes how particles interact at low energies (like the footprints). This theory is filled with numbers called Wilson coefficients. These numbers tell us the strength of various interactions, but they don't explicitly tell us what heavy particles exist deep in the "ultraviolet" (the high-energy, fundamental realm).

Usually, figuring out the heavy particles from these low-energy footprints is like trying to guess the entire recipe of a complex cake just by tasting a single crumb. It seems impossible because there are infinite ways to bake a cake that leaves the same crumb.

The Big Breakthrough
This paper introduces a magical new tool that allows physicists to reverse-engineer the recipe. They show that if you have a complete list of those "footprint" numbers (the Wilson coefficients), you can mathematically reconstruct the exact list of heavy particles (the spectrum) that created them.

Here is how they did it, using some simple analogies:

1. The "Logarithmic Whisper"

The authors realized that instead of looking at the messy interaction data directly, they should look at the logarithmic derivative of the data.

• Analogy: Imagine a choir singing a complex song. If you listen to the volume, it's a jumble. But if you listen to the rate of change of the volume (how fast it's getting louder or quieter), the individual voices start to separate out.
• In their math, this "rate of change" (which they call $Q(s)$) strips away all the complicated details about how strong the particles are (the residues) and leaves behind only the locations of the particles (the poles) and the gaps where no particles exist (the zeros).

2. The "Infinite Fingerprint"

Once they have this simplified "whisper" ($Q(s)$), they expand it into a long list of numbers (the coefficients $c_k$).

• Analogy: Think of these numbers as an infinite row of dominoes.
• The paper proves that if there are only a finite number of heavy particles (a finite number of dominoes), this infinite row of numbers actually has a hidden structure. It's not random; it's built from a specific, finite pattern.

3. The "Magic Matrix"

To find the particles, the authors use a clever mathematical trick involving a Hankel Matrix.

• Analogy: Imagine you take your list of numbers and arrange them into a giant grid (a matrix).
• If you keep adding more numbers to this grid, the grid eventually stops growing in complexity. It hits a "saturation point."
• The Clue: The size of the grid at the moment it stops growing tells you exactly how many particles exist.
  • If the grid stops growing at size $5 \times 5$, you know there are 5 particles (or 5 "features" like poles and zeros).

4. The "Reverse Engineer"

Once they know how many particles there are, they solve a simple equation (a characteristic equation) to find where they are.

• Analogy: It's like tuning a radio. You know there are exactly 5 stations broadcasting. By solving the equation, you get the exact frequency (the mass) of each station.
• The result is a complete map of the "Ultraviolet" world: the masses of all the heavy particles and the exact mathematical formula for how they interact.

Why This Matters

• For Finite Theories: If the universe has a finite number of heavy particles, this method is exact. You can get the perfect answer.
• For Infinite Theories (like String Theory): Even if there are infinite particles (an infinite tower of them), this method works as an approximation. By taking more and more "footprints" (more Wilson coefficients), you get closer and closer to the true spectrum, much like how a low-resolution photo becomes a high-definition image as you add more pixels.

The "Double Copy" Bonus

The paper also shows that this method works beautifully for "Double Copy" theories (where gravity is mathematically related to the square of particle physics forces). Because their method uses logarithms, it handles multiplication naturally.

• Analogy: If the first theory is a song, and the second is the same song played twice as loud, their method can easily figure out the second song just by analyzing the first.

Summary

In short, this paper solves the EFT Inverse Problem. It proves that the "low-energy shadow" cast by heavy particles contains a hidden code. By using a specific mathematical algorithm (involving logs and matrices), physicists can decode that shadow to reveal the full, high-energy cast of characters, turning a vague guess into a precise reconstruction of the fundamental laws of nature.

Technical Summary

1. The Problem: The EFT Inverse Problem

Effective Field Theories (EFTs) describe low-energy physics using an infinite series of Wilson coefficients ($c_k$) that parameterize local interactions. While the mapping from a fundamental Ultraviolet (UV) theory (a specific set of particles and interactions) to these low-energy coefficients is well-defined, the inverse problem is notoriously difficult: Can one uniquely reconstruct the UV spectrum (masses and residues of heavy particles) and the full scattering amplitude solely from the low-energy Wilson coefficients?

Traditional dispersion relations are linear in the scattering amplitude $A(s)$ and often fail in regimes where amplitudes grow exponentially (e.g., hard scattering in string theory) or when the number of degrees of freedom is infinite. This paper addresses the challenge of inverting the EFT expansion to recover the exact tree-level UV completion.

2. Methodology

The authors propose a novel algorithm based on nonlinear dispersion relations applied to the logarithmic derivative of the scattering amplitude.

A. The Central Object: The Logarithmic Derivative

Instead of working directly with the amplitude $A(s)$, the authors define a quotient function:
$$Q(s) = \frac{d}{ds} \log A(s) = \frac{A'(s)}{A(s)}$$
For a tree-level theory with a finite number of particles, $A(s)$ is a rational function. Consequently, $Q(s)$ exhibits a remarkably simple structure: it is a sum of simple poles located exactly at the poles ($\mu_n$) and zeros ($\rho_n$) of the original amplitude, with residues of $\pm 1$. Crucially, the dependence on the residues (coupling strengths) of the original amplitude cancels out in $Q(s)$, leaving only the locations of the singularities.

$$Q(s) = -\sum_{\text{poles}} \frac{1}{s - \mu_n} + \sum_{\text{zeros}} \frac{1}{s - \rho_n} + \text{entire terms}$$

B. Connection to Wilson Coefficients

The low-energy expansion of $Q(s)$ is defined as:
$$Q(s) = \sum_{k} c_k s^k$$
Using a contour integral around the origin in the complex $s$-plane, the authors derive a generalized Cauchy argument principle relating these coefficients $c_k$ directly to the locations of the poles and zeros:
$$c_k = \sum_{\text{poles}} \frac{1}{\mu_n^{1+k}} - \sum_{\text{zeros}} \frac{1}{\rho_n^{1+k}}$$
This establishes that the Wilson coefficients are essentially moment sums over the inverse locations of the UV spectrum.

C. The Reconstruction Algorithm (Spectroscopy)

The authors demonstrate that the sequence of coefficients $c_k$ can be organized into a Hankel matrix $c_r$:
$$[c_r]_{ij} = c_{r+i+j}$$
If the theory has a finite number of poles and zeros ($d = d_{\text{poles}} + d_{\text{zeros}}$), this infinite matrix has a finite rank $d$. The reconstruction proceeds in four steps:

1. Rank Determination: Compute the rank of the leading principal submatrices $c_r^{(\ell)}$. The rank stabilizes at $d$ when $\ell = d$. The vanishing of the determinant $\det(c_r^{(\ell)})$ signals that $\ell = d+1$, revealing the total number of states.
2. Eigenvalue Extraction: Solve the generalized eigenvalue problem:
   $$\det(c_r^{(d)} - \lambda c_{r+1}^{(d)}) = 0$$
   The eigenvalues $\lambda_n$ correspond to the inverse locations of the poles and zeros ($1/\mu_n$ and $1/\rho_n$).
3. Sign Determination: By analyzing the diagonal elements of the transformed matrix, one can distinguish whether a specific $1/\lambda_n$ corresponds to a pole ($\sigma_n = +1$) or a zero ($\sigma_n = -1$).
4. Full Reconstruction: Once the locations are known, the full functional form of the amplitude $A(s)$ is reconstructed.

3. Key Contributions and Results

• Exact Inversion for Finite Spectra: The paper proves that for any tree-level QFT with a finite number of particles, the UV spectrum is exactly encoded in a finite number of Wilson coefficients. The algorithm provides a deterministic way to extract masses and zeros without ambiguity.
• Nonlinear Dispersion Relations: The authors introduce a new class of dispersion relations that are nonlinear in the amplitude (via the log derivative). This allows the method to bypass the boundary term issues that plague conventional linear dispersion relations in theories with exponential growth (e.g., string theory).
• String Theory Application: The method is successfully applied to the Veneziano amplitude (open strings) and the Virasoro-Shapiro amplitude (closed strings).
  • Even though string theory has an infinite tower of particles ($d = \infty$), the authors show that the truncated Hankel matrix approach converges to the exact spectrum via Padé approximants.
  • They demonstrate that the Wilson coefficients of the Veneziano amplitude (expressed as multiple zeta values) can be inverted to recover the infinite Regge trajectory of string states.
• Double Copy Relations: The logarithmic nature of $Q(s)$ makes it compatible with the "Double Copy" structure (KLT relations), where closed string amplitudes are products of open string amplitudes. The authors derive a relation showing how the Wilson coefficients of closed strings are sums of those from open strings plus corrections from the KLT kernel.

4. Significance and Implications

• Bridging IR and UV: This work provides a rigorous mathematical framework for "UV reconstruction" from IR data, moving beyond positivity bounds to full spectral reconstruction.
• Robustness: The method is insensitive to the high-energy behavior of the amplitude (as long as it is polynomially bounded in the log), making it applicable to theories where standard dispersion relations fail.
• Experimental Relevance: While currently theoretical, the paper outlines how experimental measurements of low-energy scattering (determining $c_k$) could theoretically be used to predict the existence and masses of heavy resonances or string excitations, even if they are far beyond the direct energy reach of the collider.
• Limitations: The current framework is restricted to tree-level amplitudes. Loop corrections introduce branch cuts and logarithms that break the meromorphic structure required for the simple pole/zero decomposition. Extending this to loop level remains an open challenge.

Conclusion

Calisto et al. have solved the "EFT inverse problem" for tree-level theories by transforming the problem of spectral reconstruction into a linear algebra problem involving Hankel matrices of Wilson coefficients. This approach unifies the study of EFTs, moment problems, and Padé approximation, offering a powerful tool to decode the hidden ultraviolet structure of quantum field theories and string theory directly from low-energy data.