AI 中文总结
针对脉冲点云模型全微调开销大、亚阈值信息被抑制的问题，提出首个参数高效微调框架SpikePEFT，通过IDT和SSDA技术，在仅更新约5%参数的情况下，在多个基准上取得优异分类准确率，同时保持SNN的节能特性。
AI 中文摘要
脉冲神经网络（SNN）通过事件驱动计算为资源受限设备上的点云分析提供节能解决方案。然而，现有的预训练脉冲点云模型依赖全微调进行下游任务适配，会产生大量参数和存储开销；此外，二值脉冲传播会抑制任务相关的亚阈值信息。为解决这些问题，我们提出SpikePEFT，首个针对脉冲点云模型的参数高效微调框架。具体而言，固有动力学调优（IDT）自适应调节膜衰减和发放阈值，在保持预训练突触变换冻结的同时实现高效的神经元固有适配；此外，静默态消歧适配（SSDA）从具有信息价值的静默态中恢复任务相关信息，从而为下游适配提供更丰富的证据。在多个基准上的大量实验证明了SpikePEFT的有效性和效率，特别是，我们的方法在ModelNet40上达到92.4%的准确率，在最具挑战性的分类拆分ScanObjectNN(PB_T50_RS)上达到85.6%，同时仅更新约5%的可训练参数并保持SNN的节能特性。本研究为神经形态视觉模型的参数高效适配迈出了重要一步。
英文摘要
Spiking Neural Networks (SNNs) offer energy-efficient solutions for point cloud analysis on resource-constrained devices through event-driven computation. However, existing pre-trained spiking point cloud models rely on full fine-tuning for downstream task adaptation, incurring substantial parameter and storage overhead. Furthermore, binary spike propagation suppresses task-relevant sub-threshold information. To address these issues, we propose SpikePEFT, the first parameter-efficient fine-tuning framework for spiking point cloud models. Specifically, Intrinsic Dynamics Tuning (IDT) adaptively modulates membrane decay and firing thresholds, enabling efficient neuron-intrinsic adaptation while keeping the pre-trained synaptic transformations frozen. Moreover, Silent-State Disambiguation Adaptation (SSDA) recovers task-relevant information from informative silent states, thereby providing richer evidence for downstream adaptation. Extensive experiments across multiple benchmarks demonstrate the effectiveness and efficiency of SpikePEFT. In particular, our method achieves 92.4% accuracy on ModelNet40 and 85.6\% on the most challenging classification split ScanObjectNN(PB\_T50\_RS) while updating only about 5% of the trainable parameters and preserving the energy efficiency of SNNs. This work provides a promising step toward parameter-efficient adaptation of neuromorphic vision models.
