Nouhaila Innan

LLM-generated quantum tests are executable but fail to reliably distinguish correct from faulty code, with the best models achieving only 71% and 34% test quality on separate benchmarks.

Physical noise can enhance the performance of quantum neural networks, but its benefits are not universal across all datasets.

Confidence-aware selective recovery can drastically improve logical-level performance in quantum error correction, outperforming traditional unconditional teacher replacements.

Hybrid quantum algorithms can be sped up by almost 2 orders of magnitude simply by co-scheduling classical and quantum resources in a stateful, calibration-aware manner.

General-purpose LLMs can't hack it in quantum coding, but a RAG system armed with a code-aware embedding strategy can boost performance by 28% on real-world quantum programming challenges.

Quantum-inspired node embeddings can outperform classical baselines in graph neural networks, but only on structure-driven datasets, revealing critical dataset dependencies.

Turns out, simpler quantum circuits with the right data encoding can outperform complex ones in diagnosing Chronic Kidney Disease.