QuPID: Quantum Parameter-Efficient Input-Dependent Retrieval Adaptation for Medical RAG

2026-09-27

AI Summary

This study addresses the bottleneck in quantum retrieval where shared unitary transformations yield constant fidelity and untrainable rankings, proposing the QuPID framework. This method pioneers an input-dependent quantum retrieval adaptation mechanism based on local Pauli expectation values. By leveraging data re-uploading to construct a structured factorized quadratic feature map, QuPID adapts to medical image features by comparing measurement readouts rather than quantum states. Experimental results demonstrate that with merely 60 parameters, QuPID significantly outperforms million-parameter LoRA adapters and frozen encoders in terms of P@5 on datasets such as ChestX-ray14. This work achieves a performance breakthrough under extreme parameter efficiency for quantum-enhanced medical image retrieval.

Abstract

Fidelity-based quantum retrieval ranks candidates by the fidelity between query and archive states. Applying a shared input-independent unitary after fixed state encoding leaves that fidelity unchanged, so training the circuit cannot alter the ranking. Quantum parameter-efficient input-dependent retrieval adaptation (QuPID) repairs this by making the circuit input-dependent through data re-uploading and by comparing measurement readouts, vectors of local Pauli expectations, rather than states. The result is a small readout for adapting frozen image features to a local archive with limited data: training simulates the circuit classically, and inference runs on a GPU with fixed learned parameters. We characterize the class as a structured factorization of input-modulated quadratic feature maps, bound the frequency support of its re-uploading channel, and give a parameter-count generalization bound that motivates its small budget. Under a shared frozen backbone and a label-free protocol, QuPID's 60 parameters give higher precision-at-5 (P@5) on ChestX-ray14 and MURA than frozen medical encoders, and than adapters and low-rank adaptation (LoRA) with up to 5.25 million trainable parameters. On ChestX-ray14, the P@5 gain over the frozen encoder is +0.116, the lead over retuned adapters is widest at 512 adaptation examples (+0.040), and the full-budget margin over an equally compact classical rotation-plane head is +0.023 with a 95% interval excluding zero. Medical imaging is the primary testbed; the pattern recurs on two non-medical benchmarks, in report generation, and under simulated gate noise and finite-shot readout.

Problem

• Quantum Retrieval
• Medical RAG
• Fidelity-based Ranking
• Parameter-Efficient Adaptation

Innovation

• Quantum Retrieval
• Data Re-uploading
• Parameter-Efficient Adaptation
• Medical RAG
• Pauli Expectations