| name | quantum-software-testing-benchmark |
| category | quantum |
| description | Qolumbina benchmark infrastructure for quantum software testing (QST) — curates 40 scalable quantum programs from open-source repos with systematic selection, refactoring, specifications, unit tests, and standardized interfaces. Introduces QST-oriented criteria along functionality, output behavior, development complexity, and quantum-specific execution complexity. |
| tags | ["quantum","software-testing","benchmarking","QST","software-engineering"] |
| arxiv_id | 2607.02029v1 |
| created | 2026-07-07 |
Quantum Software Testing Benchmark (Qolumbina)
Overview
Qolumbina addresses a critical gap in quantum software testing (QST) research: existing studies rely on small hard-coded or circuit-level benchmarks, while available quantum programs are scattered across repositories without clear selection criteria. Qolumbina provides a controlled, scalable benchmark infrastructure for rigorous empirical QST evaluation.
Core Contribution
Program Curation Pipeline
- Collection: Gather 40 programs from open-source quantum repositories
- Selection: Apply systematic criteria for test-relevance
- Refactoring: Standardize code structure and interfaces
- Specifications: Define expected behavior for each program
- Test Cases: Create unit test examples with known outcomes
- Standardization: Provide uniform interfaces for testing tools
QST-Oriented Characterization Criteria
Quantum programs are characterized along four dimensions:
- Functionality: What the program computes (e.g., optimization, simulation, ML)
- Output Behavior: Deterministic vs probabilistic, scalar vs distribution
- Development Complexity: Code size, dependency depth, language constructs used
- Quantum-Specific Execution Complexity: Qubit count, circuit depth, gate types, backend requirements
Scalability Analysis
- Supports scalability analysis beyond fixed-size circuit benchmarks
- Programs can be parameterized to test at different scales
- Enables studying how testing approaches perform as programs grow
Empirical Findings
Benchmark Coverage
Qolumbina covers diverse testing-relevant properties including:
- Multiple quantum programming frameworks
- Various algorithm categories (VQA, QAOA, quantum ML, etc.)
- Different circuit depths and qubit counts
- Both simulation-ready and hardware-targetable programs
Testing Approach Evaluation
Through controlled experiments with two recent QST approaches:
- Demonstrated feasibility for execution-cost studies
- Demonstrated feasibility for fault-detection studies
- Identified backend-dependent effects that can influence QST result interpretation
Implementation Pattern
Setting Up a QST Experiment
1. Select programs from Qolumbina matching your testing criteria
2. Define the testing approach/mutation strategy
3. Run tests across the benchmark suite
4. Measure: fault detection rate, execution cost, false positives
5. Analyze backend-dependent effects on results
6. Report results with locked audit scope (see CLAIMSTAB-QC skill)
Integration with CLAIMSTAB-QC
For auditing empirical comparisons between testing approaches:
- Record baselines, metrics, and admissible evidence
- Lock comparison design before computing outcomes
- Classify results as Sustained, Unresolved, or Reversed
When to Use
- Developing new quantum software testing approaches
- Benchmarking existing QST tools
- Studying scalability of quantum testing methods
- Comparing quantum testing approaches fairly
- Understanding backend effects on test results
Activation Keywords
quantum software testing, QST, Qolumbina, quantum benchmark, quantum program testing, quantum test infrastructure, quantum testing scalability, backend-dependent testing, quantum mutation testing