Resistive In-Memory Computing (RIMC) offers ultra-efficient computation for edge AI but faces accuracy degradation due to RRAM conductance drift over time. Traditional retraining methods are limited by RRAM's high energy consumption, write latency, and endurance constraints. We propose a DoRA-based calibration framework that restores accuracy by compensating influential weights with minimal calibration parameters stored in SRAM, leaving RRAM weights untouched. This eliminates in-field RRAM writes, ensuring energy-efficient, fast, and reliable calibration. Experiments on RIMC-based ResNet50 (ImageNet-1K) demonstrate 69.53% accuracy restoration using just 10 calibration samples while updating only 2.34% of parameters.

翻译：阻变式存内计算为边缘AI提供了超高能效的计算能力，但其精度会因RRAM电导随时间漂移而下降。传统重训练方法受限于RRAM的高能耗、写入延迟和耐久性约束。本文提出一种基于DoRA的校准框架，通过将关键权重补偿信息以最小化校准参数形式存储于SRAM中（保持RRAM权重不变），从而恢复计算精度。该方法避免了现场RRAM写入操作，实现了高能效、快速且可靠的校准。在基于RRAM存内计算的ResNet50（ImageNet-1K）上的实验表明，仅使用10个校准样本并更新2.34%的参数即可实现69.53%的精度恢复。