Submission Process

The Quantum Advantage Tracker (QAT) relies on community contributions to build a comprehensive map of quantum computing progress. This section outlines how researchers and developers can submit results for existing benchmarks or propose new problem instances.

Submission Overview

The submission process is handled entirely through GitHub. Contributors provide experimental results, error analysis, and institutional metadata via structured GitHub Issue Templates. Once a submission is reviewed and validated, it is merged into the project's data layer and automatically displayed on the website.

Step 1: Select Your Pathway

Before submitting, identify which of the three pathways best fits your experimental result:

| Pathway | Focus | Requirement | | :--- | :--- | :--- | | Observable Estimations | Physical property prediction | Expectation values with mathematically grounded error bars. | | Variational Problems | Optimization and Chemistry | Solution bounds (e.g., ground-state energy) that respect the variational principle. | | Classically Verifiable | Computational complexity | Quantum outputs that can be efficiently checked against classical witnesses or known answers. |

Step 2: Choose or Define an Instance

You can submit results for an existing problem instance or propose a new one.

Using Existing Instances

Navigate to the data directory to find predefined circuits and Hamiltonians:

• Observable Estimations/CVP: Look in circuit-models/ for .json or circuit specification files.
• Variational Problems: Look in hamiltonians/ for problem definitions.

Submitting Custom Models

If your experiment uses a new circuit model or Hamiltonian not currently listed, you must include the model specification in your submission. Refer to the Data Documentation for schema requirements.

Step 3: Open a Submission Ticket

Submissions are initiated by opening a new issue using a specific template. These templates ensure all necessary scientific metadata is captured.

1. Navigate to the New Issue page.
2. Select the template corresponding to your pathway:
   • 01-submission-path-observable-estimations.yml
   • 02-submission-path-variational-problems.yml
   • 03-submission-path-classically-verifiable-problems.yml
3. Fill out the required fields, including:
   • Result Data: The primary numerical output (e.g., energy value, expectation value).
   • Error Bars: For observable estimations, provide the rigorous confidence intervals.
   • Hardware/Simulator Info: Details on the quantum processor or classical simulator used.
   • Institutional Metadata: A comma-separated list of participating organizations (e.g., "University of Tokyo, IBM Quantum").

[!TIP] Do not worry about labels, assignees, or milestones. The QAT maintainers will handle categorization and tagging once the issue is submitted.

Step 4: Validation and Publication

Once an issue is submitted, the following lifecycle occurs:

1. Validation: Maintainers and community members review the submission for scientific rigor and compliance with the pathway's requirements.
2. Data Integration: Upon approval, the submission data is added to the corresponding JSON files in the repository:
   • data/observable-estimations/submissions.json
   • data/variational-problems/submissions.json
   • data/classically-verifiable-problems/submissions.json
3. Deployment: The website is automatically rebuilt. Your submission will appear in the Advantage Trackers table, and your institution will be added to the contributors' marquee on the homepage.

Contribution Format Example

When preparing your submission, ensure your data aligns with the project's internal types. For example, a circuit instance submission should correspond to the following structure:

// Internal representation for circuit-based instances
export type CircuitInstance = {
  id: string;      // Unique identifier for the instance
  path: string;    // Link to the circuit file
  qubits: number;  // Number of qubits involved
  gates: number;   // Total gate count
  type: string;    // Category/Model type
};

For detailed technical specs on data formats, please consult the src/types/ directory in the source code.