🔢 mathematics

Scaling limit of the colored ASEP and stochastic six-vertex models

This paper establishes that the height functions of colored asymmetric simple exclusion processes and stochastic six-vertex models with fully packed initial conditions converge to the Airy sheet under KPZ scaling, a result derived from a colored Hall-Littlewood line ensemble embedding that subsequently yields corollaries regarding the convergence to the directed landscape, KPZ fixed points, and the stationary horizon.

Original author: Amol Aggarwal, Ivan Corwin, Milind Hegde

Published 2026-11-06

📖 4 min read🧠 Deep dive

CC BY 4.0

Original author: Amol Aggarwal, Ivan Corwin, Milind Hegde

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the author. For technical accuracy, refer to the original paper. Read full disclaimer

In the study of how complex systems behave, scientists often look for patterns that emerge when countless tiny parts interact. Imagine a crowded room where people move around, bumping into one another, trying to get to the other side. If you watch just one person, their path seems random and chaotic. But if you step back and look at the whole crowd, a surprising order often appears. This is the realm of statistical physics, where researchers study how simple rules governing individual particles lead to large-scale structures. For decades, a specific set of rules known as the Kardar-Parisi-Zhang scaling has been the gold standard for describing how these systems fluctuate over time. It predicts that if you zoom out far enough, the messy details of individual movements smooth out into a universal shape, much like how the rough surface of a sand dune looks smooth from a distance. This theory has been confirmed for many simple systems, but it has remained unproven for more complex, multi-layered versions where particles have different "priorities" or colors, making their interactions far more intricate.

A team of mathematicians has now bridged this gap, proving that these complex, multi-colored systems do indeed settle into the same universal shape predicted by the theory. They focused on two specific models: one where particles hop along a line and another where they move through a grid of intersections. In these models, every particle is assigned a color, which acts as a priority level; a particle with a higher color can push a particle with a lower color out of the way, but not the other way around. The researchers wanted to know what happens when these systems run for a very long time and are viewed from a great distance. They found that the collective height of these colored particles, when measured with the right scale, converges to a specific, random surface known as the Airy sheet. This surface is a fundamental object in mathematics that describes the ultimate limit of how these fluctuations behave, acting as a kind of master blueprint for a wide class of random growth processes.

To reach this conclusion, the team had to navigate a landscape of mathematical complexity that had previously blocked progress. They began by embedding their colored particle systems into a larger, more structured framework called a line ensemble. You can think of this as taking the chaotic movement of the particles and organizing them into a series of parallel paths that never cross each other. These paths are governed by a set of local rules that determine their shape, similar to how a river follows the contours of the land. The researchers showed that the behavior of the entire system is largely determined by the topmost path in this ensemble. By carefully analyzing how these paths interact and separate from one another, they proved that the top path settles into a specific, wavy shape known as the Airy line ensemble. Once this foundation was established, they demonstrated that the other paths in the system are essentially determined by the top one, creating a unified structure that matches the Airy sheet.

The implications of this discovery extend far beyond the specific models they studied. Because the colored systems they analyzed are deeply connected to other well-known models in physics and probability, their findings act as a key that unlocks the behavior of many related systems. For instance, they showed that when these colored systems start from different initial arrangements, their long-term evolution converges to a structure called the directed landscape. This landscape is a four-dimensional object that describes how information and fluctuations travel through space and time in these systems. Furthermore, they proved that the stationary states of these systems—what they look like after running for an infinite amount of time—converge to a limit known as the stationary horizon. This confirms a long-standing prediction that these complex, multi-component systems share the same ultimate fate as their simpler, single-color counterparts.

One of the most striking aspects of their work is the demonstration of decoupling. In these systems, particles moving along different paths usually influence each other. However, the researchers showed that if two groups of particles are moving along paths that are sufficiently far apart, they eventually stop affecting one another. Their fluctuations become independent, as if they were in separate universes. This behavior was predicted by theories of non-linear fluctuating hydrodynamics, but this paper provides the first rigorous proof for these specific colored models. By proving that the complex interactions between different colored particles simplify into these universal limits, the authors have provided a complete picture of how these systems behave at the largest scales. Their work confirms that even in a world of intricate, multi-layered interactions, nature tends to simplify into a few elegant, universal forms.

Technical Summary: Scaling Limit of the Colored ASEP and S6V Models

1. Problem Statement

This paper addresses the limiting fluctuations of two fundamental integrable probability models: the colored asymmetric simple exclusion process (colored ASEP) and the colored stochastic six-vertex (colored S6V) model. While the scaling limits of their uncolored (single-species) counterparts are well-understood—converging to the KPZ fixed point and related objects like the Airy line ensemble and directed landscape—the behavior of multi-species (colored) systems under the Kardar-Parisi-Zhang (KPZ) scaling regime has remained largely open.

Specifically, the authors investigate the joint scaling limit of the height functions for these models under packed initial conditions (where particles of color $-k$ occupy site $k$ for all $k \in \mathbb{Z}$ ). The central question is whether these multi-species height functions converge to the Airy sheet and, more generally, the directed landscape, which are the conjectured universal scaling limits for the KPZ universality class in two dimensions.

2. Methodology

The authors develop a novel framework to analyze the edge scaling limit of these colored models, bypassing the need for an exact "dual" representation (such as the Robinson-Schensted-Knuth correspondence used in solvable last passage percolation models) which is not known to exist for colored ASEP/S6V.

2.1. Embedding into Colored Line Ensembles

The core methodological innovation is the embedding of the colored S6V model into a larger structure called the colored Hall-Littlewood line ensemble.

Connection: Using the Yang-Baxter equation, the authors establish a distributional match between the colored S6V model and a colored $q$ -Boson model on a negative quadrant.
Line Ensemble Construction: The height function of the colored S6V model is identified with the top curves of a family of $n$ coupled line ensembles (indexed by color), where each line ensemble consists of non-intersecting Bernoulli paths. This structure possesses a colored Gibbs property, meaning the conditional law of paths in an interval depends explicitly on the boundary data and the paths outside the interval via specific Boltzmann weights.

2.2. Analysis of the Gibbs Property

A major technical hurdle is that these colored Gibbsian line ensembles do not satisfy the Fortuin-Kasteleyn-Ginibre (FKG) inequality (stochastic monotonicity), a property heavily relied upon in previous tightness proofs for uncolored line ensembles (e.g., Corwin-Hammond).

Overcoming Lack of Monotonicity: The authors develop a new framework to prove tightness and separation of the curves without monotonicity. They utilize a weak monotonicity property and establish a priori separation estimates between curves. By showing that curves are likely separated by a distance of order $N^{1/3}$ (where $N$ is the scaling parameter), they demonstrate that the reweighting factors in the Gibbs property (which involve powers of $q$ ) become negligible, effectively reducing the system to one resembling non-intersecting Brownian bridges.
Approximate LPP Representation: The authors prove that the coupled colored line ensembles can be approximated by a Last Passage Percolation (LPP) problem across a single "determining" line ensemble (the one indexed by the minimal color). This serves as a substitute for the exact dual LPP representation used in other KPZ models.

2.3. Convergence Strategy

The proof proceeds in two main stages:

Uncolored Limit: They first prove that the "determining" uncolored Hall-Littlewood line ensemble converges to the parabolic Airy line ensemble. This involves establishing tightness and showing that any subsequential limit satisfies the Brownian Gibbs property, which uniquely characterizes the Airy line ensemble.
Colored Limit: Using the approximate LPP representation, they show that the remaining colored line ensembles converge to limits determined by LPP problems across the parabolic Airy line ensemble. By the definition of the Airy sheet (as an LPP across the Airy line ensemble), this implies the convergence of the full colored system to the Airy sheet.

3. Key Contributions and Results

3.1. Convergence to the Airy Sheet

The primary result is the proof that the rescaled height functions of the colored ASEP and colored S6V models, under packed initial conditions, converge weakly to the Airy sheet $S(x, y)$ .

Theorem 2.8 (ASEP): The rescaled colored ASEP height function converges to the Airy sheet.
Theorem 2.20 (S6V): The rescaled colored S6V height function converges to the Airy sheet.
This is the first proof of convergence to the Airy sheet for any colored interacting particle system.

3.2. Convergence to the Directed Landscape and KPZ Fixed Point

As corollaries, the authors establish:

Directed Landscape: The space-time scaling limit of the colored models (under basic coupling) converges to the directed landscape (Theorem 2.12 and 2.23).
Coupled KPZ Fixed Points: For multiple general initial conditions coupled via the basic coupling, the height functions converge to KPZ fixed points coupled through the directed landscape. This result is detailed in a follow-up paper [ACH24] but is stated here as a consequence of the Airy sheet convergence.

3.3. Stationary Horizon and Decoupling

Stationary Horizon: The paper proves that the stationary measures of the colored ASEP (and S6V) converge to the stationary horizon (Corollary 2.14), a process introduced by Busani [Bus24] as the universal limit of multi-type invariant distributions.
Decoupling: The authors prove a strong form of decoupling for the colored height functions and stationary two-point functions. They show that if the characteristic slopes of two components differ by an amount growing slightly faster than the KPZ scaling ( $\beta \gg \epsilon^{1/3}$ ), the fluctuations become asymptotically independent. This verifies predictions from the theory of non-linear fluctuating hydrodynamics (Corollaries 2.15 and 2.16).

4. Significance and Claims

The paper claims to provide the first rigorous proof of the Airy sheet and directed landscape scaling limits for colored interacting particle systems. Its significance lies in:

Bridging the Gap: It resolves the open problem of identifying the universal scaling limit for colored ASEP and S6V, which were previously only understood through marginal matching or specific one-point results.
Methodological Advancement: It introduces a robust framework for analyzing colored Gibbsian line ensembles that lack stochastic monotonicity. This framework, relying on approximate LPP representations and separation estimates, is presented as a blueprint for analyzing other colored integrable models where exact dual representations are unavailable.
Universality Verification: It confirms that the colored models belong to the same universality class as uncolored models and Brownian LPP, converging to the same limiting objects (Airy sheet, directed landscape, stationary horizon).

The authors explicitly state that their work does not rely on an exact dual LPP representation for the colored models (which is not expected to exist) but rather on an approximate representation derived from the Yang-Baxter equation and the structure of the colored Hall-Littlewood line ensemble. This approach allows them to access the edge scaling limit without the algebraic solvability that typically underpins such results in the uncolored case.