Quantum-Inspired Spectral Geometry for Neural Operator Equivalence and Structured Pruning
This addresses operator redundancy for multimodal AI on heterogeneous domestic hardware, with a novel theoretical foundation for cross-modal substitutability.
The paper tackles the problem of neural operator redundancy in multimodal AI on resource-constrained hardware by introducing a quantum-inspired geometric framework that represents operators via normalized singular value spectra on a Bloch hypersphere. It proves a spectral-to-functional equivalence theorem showing vanishing distance implies functional closeness, and proposes a metric-driven pruning method validated as superior to baselines in simulations.
The rapid growth of multimodal intelligence on resource-constrained and heterogeneous domestic hardware exposes critical bottlenecks: multimodal feature heterogeneity, real-time requirements in dynamic scenarios, and hardware-specific operator redundancy. This work introduces a quantum-inspired geometric framework for neural operators that represents each operator by its normalized singular value spectrum on the Bloch hypersphere. We prove a tight spectral-to-functional equivalence theorem showing that vanishing Fubini--Study/Wasserstein-2 distance implies provable functional closeness, establishing the first rigorous foundation for cross-modal and cross-architecture operator substitutability. Based on this metric, we propose Quantum Metric-Driven Functional Redundancy Graphs (QM-FRG) and one-shot structured pruning. Controlled simulation validates the superiority of the proposed metric over magnitude and random baselines. An extensive experimental validation on large-scale multimodal transformers and domestic heterogeneous hardware (Huawei Ascend, Cambricon MLU, Kunlunxin) hardware is deferred to an extended journal version currently in preparation.