⚛️ quantum physics

Constant-depth adaptive preparation of Dicke and symmetric states

This paper introduces an exact, constant-depth adaptive protocol for preparing arbitrary Dicke and permutation-symmetric states using polynomial ancillary qubits, leveraging measurements and classical feedforward to achieve high success probabilities without increasing circuit depth.

Original authors: Rui Luo, Junjie Chen, Xiongfeng Ma

Published 2026-08-04

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Rui Luo, Junjie Chen, Xiongfeng Ma

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 authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine a world where computers don't just crunch numbers but dance with the very fabric of reality. This is the realm of quantum computing, a field where information isn't stored in simple on/off switches (bits) but in "qubits" that can be in multiple states at once, like a spinning coin that is both heads and tails until you catch it. One of the biggest challenges in this world is getting these qubits to work together in perfect harmony. Scientists often need to create specific, highly entangled patterns of qubits called "Dicke states." Think of these states as a perfectly choreographed dance troupe where, no matter how you shuffle the dancers, the group looks exactly the same. These patterns are crucial for super-sensitive measurements (like detecting tiny gravitational waves) and building future quantum networks. However, creating these dances usually takes a long time and a lot of complex steps, which is a problem because quantum systems are fragile and lose their "magic" (coherence) if you wait too long. The goal is to create these states as quickly as possible, ideally in a single, instant step, without breaking the delicate quantum rules.

Enter a new study by Rui Luo, Junjie Chen, and Xiongfeng Ma from Tsinghua University, which offers a clever, lightning-fast way to organize these quantum dancers. The researchers have developed a method to prepare these complex quantum states in "constant depth," meaning the time it takes doesn't get longer even if you add more dancers to the troupe. They achieved this by using a trick called "adaptive preparation," which is like having a conductor who can listen to the orchestra mid-performance and instantly tell the musicians how to adjust their notes to get the perfect sound, rather than just following a rigid, pre-written score.

The core of their discovery is a new way to build a "Uniform Subset Superposition" (USS). Imagine you have a giant bag of numbered balls, and you want to pull out a specific number of them, but you want every possible combination to be equally likely. Usually, sorting these balls takes time. The authors found a way to do this sorting instantly using a mix of quantum magic and quick classical math. They first create a messy pile of random numbers, then use a "sorting hat" (their adaptive circuit) to instantly organize them into a neat, ordered line. A single attempt at this sorting succeeds with a probability of at least 1/k. If the sorting fails, they just try again immediately; because they can repeat this process in parallel, the chance of failure drops to almost zero without slowing down the process.

The paper shows that this method can prepare these states exactly, with a high success rate after parallel repetition, using a manageable number of extra "helper" qubits (ancillae). Specifically, for a system of $n$ qubits with $k$ excitations, their method uses roughly $O(n^2 + k^2 \log^2 n + kn \log n \log \log n)$ extra qubits. While this might sound like a lot, it's a significant improvement over previous methods for certain sizes, and it guarantees the result is perfect, not just an approximation.

Furthermore, the authors didn't just stop at one type of dance. They built a "lifting framework," which is like a universal adapter. They showed that if you have a way to make any single type of Dicke state (a specific dance move), you can use their framework to combine them to create any symmetric state (any complex choreography) in the same amount of time. To achieve this for arbitrary symmetric states, they combined their framework with a recent, separate constant-depth unitary method (a different kind of quantum dance that doesn't rely on their adaptive sorting trick) to create a recipe to prepare any symmetric state of $n$ qubits using $O(n^3 \sqrt{\log n})$ extra qubits, all in constant time.

The paper is careful to note that while their method is exact and fast, it relies on the ability to measure qubits and feed that information back instantly, a feature available in some advanced quantum setups but not all. They also point out that while their resource usage is efficient, it's not yet the absolute minimum possible, leaving room for future improvements. However, their work provides a powerful new toolkit for quantum engineers, turning a slow, difficult process into a rapid, reliable one, paving the way for more complex quantum simulations and networks.

Technical Summary: Constant-Depth Adaptive Preparation of Dicke and Symmetric States

Problem Statement
Efficient preparation of Dicke states ( $|D_n^k\rangle$ ) and general permutation-symmetric states is a fundamental primitive for quantum metrology, networking, and collective information processing. However, preparing these states with low quantum depth is challenging. In standard unitary circuit models with bounded-fan-in gates, generating long-range correlations typically requires non-constant depth, with logarithmic lower bounds established for representative entangled states. While intermediate measurements and classical feedforward (adaptive circuits) can circumvent light-cone restrictions to achieve constant depth, existing protocols often rely on deterministic Grover-search-based filtering or incur significant resource overheads. The authors aim to develop an exact, constant-depth adaptive protocol for arbitrary Dicke states and, subsequently, arbitrary symmetric states, optimizing the trade-off between circuit depth and ancillary qubit count.

Methodology
The authors propose a two-stage approach: first, constructing a constant-depth adaptive protocol for Dicke states using a "Uniform Subset Superposition" (USS) primitive; second, establishing a general "lifting framework" to extend unitary Dicke-state preparation to arbitrary symmetric states.

Uniform Subset Superposition (USS) Preparation:
The core of the Dicke-state protocol is the preparation of the USS state, defined as $|USS_{n,k}\rangle = \binom{n}{k}^{-1/2} \sum_{0 \le j_1 < \dots < j_k \le n-1} |j_1\rangle \dots |j_k\rangle$ .
- Adaptive Ordering: The protocol begins by preparing $k$ independent uniform index registers. It then employs a constant-depth adaptive circuit to compare these indices and determine their relative order.
- Measurement and Classical Processing: Intermediate measurements are performed on the comparison outcomes to determine the "descent number" of the permutation of indices. Based on these classical results, a "lifting" operation is applied to convert weak inequalities (allowing equal indices) into strict inequalities, ensuring the indices are distinct and ordered.
- Success Probability: A single execution succeeds with a probability of at least $1/k$ . The failure probability is suppressed exponentially via parallel repetition without increasing the quantum depth.
- Conversion to Dicke State: Once $|USS_{n,k}\rangle$ is prepared, it serves as an index register. The protocol writes excitations into a target register at the positions specified by the indices and then coherently erases the index register, leaving the target in the state $|D_n^k\rangle$ .
Lifting Framework for Symmetric States:
To prepare arbitrary symmetric states ( $|\psi\rangle = \sum a_k |D_n^k\rangle$ ), the authors introduce a framework that coherently combines clean unitary Dicke-state preparation circuits.
- Coherent Control: The method encodes the desired coefficients $\{a_k\}$ into a one-hot index register.
- Controlled Application: Controlled versions of the underlying unitary Dicke-state preparation circuits ( $U_{n,k}$ ) are applied in parallel, conditioned on the index register.
- Erasure: The index register is erased using Hamming-weight computation and equality checks, leaving the target register in the desired superposition.

Key Contributions and Results

Constant-Depth Dicke-State Preparation (Theorem 1):
The authors present an exact, probabilistic, constant-depth adaptive protocol for preparing $|D_n^k\rangle$ .
- Resources: It uses $O(n^2 + k^2 \log^2 n + kn \log n \log \log n)$ ancillary qubits.
- Depth: The adaptive quantum depth is $O(1)$ .
- Success: The success probability is at least $1/k$ . By parallel repetition, the failure probability can be reduced to $e^{-c}$ for any constant $c$ , with ancillary overhead scaling linearly with $c$ .
- Comparison: This improves upon previous adaptive constructions (e.g., Ref. [30]) for arbitrary $k$ and offers a mechanism distinct from the unitary Grover-search approach used in Ref. [36].
Uniform Subset Superposition as a Primitive:
The paper establishes that $|USS_{n,k}\rangle$ can be prepared exactly in constant depth with $O(k^2 \log^2 n)$ ancillary qubits. The authors note that this state is of independent interest as the uniform vertex state of the Johnson graph (relevant for quantum walks and topological data analysis).
General Lifting Framework (Theorem 2):
The authors prove that if a family of clean unitary circuits prepares $|D_n^k\rangle$ with depth $O(L)$ and ancillary cost $O(\xi)$ , then any arbitrary $n$ -qubit symmetric state can be prepared with depth $O(L)$ and ancillary cost $O(n^2 \log n + n\xi)$ . This separates the complexity of combining Hamming-weight sectors from the complexity of preparing individual Dicke states.
Constant-Depth Symmetric-State Preparation (Corollary 3):
By combining the lifting framework with recent constant-depth unitary Dicke-state constructions (Ref. [36]), the authors derive an exact constant-depth protocol for arbitrary symmetric states using $O(n^3 \sqrt{\log n})$ ancillary qubits.

Significance and Claims
The paper claims to provide the first exact constant-depth adaptive protocol for arbitrary Dicke states that avoids deterministic Grover search, utilizing a fundamentally different mechanism based on adaptive ordering and measurement. The work demonstrates that measurements and classical feedforward can effectively replace deep unitary correlations for state preparation.

Regarding symmetric states, the authors claim that their lifting framework provides a general reduction: any improvement in the resource complexity of unitary Dicke-state preparation immediately translates to an improvement for general symmetric states. They assert that their resulting protocol for arbitrary symmetric states ( $O(n^3 \sqrt{\log n})$ ancillas) represents the best known asymptotic ancillary-qubit complexity for constant-depth exact preparation.

The authors remain modest regarding optimality, noting that the lower bound for ancillary qubits is $\Omega(n)$ , leaving a gap between their $O(n^2)$ -scale usage and the theoretical minimum. They also highlight that their adaptive protocol relies on intermediate measurements, which prevents its direct use as a controlled subroutine in the lifting framework (which requires clean unitary circuits), necessitating the combination with the unitary construction from Ref. [36].

Technical Summary: Constant-Depth Adaptive Preparation of Dicke and Symmetric States

More like this