Quantum-Assisted Data Foundry
Physical-QPU state selection connected to executable software verification.
Q-Psi demonstrated a physical-QPU-assisted coding-data pipeline in which QPU measurement samples causally selected repository mutation states that were converted by executable software verifiers into 93 unique verified repair records.
The experiment demonstrates mechanical pipeline feasibility, not improved model quality or quantum advantage. Downstream LLM fine-tuning was not evaluated.
Data Foundry End-to-End Pipeline
Classical Verification Throughput
Candidate-state supply exceeded current classical verifier throughput. In the 120-Qsec campaign, the physical QPU supplied 65,138 unique candidate states within 127.58 seconds of quantum execution, while classical test execution sandboxes evaluated 158 candidates in 246.3 seconds.
What Was NOT Tested
The experiment demonstrates pipeline feasibility, not improved model quality or quantum advantage. Downstream LLM fine-tuning was not evaluated.
No claim is made that quantum-selected datasets produce superior model accuracy compared to classical data synthesis, or that language models trained on these records outperform baseline models.
da1vjh4dedkc73erijq0 (83,192 shots • 31.90s)da1vjmeg52gs73cm5us0 (83,192 shots • 31.90s)da1vk3mg52gs73cm607g (83,192 shots • 31.90s)da1vke63kjvs738795og (83,192 shots • 31.88s)