# QuIDE: Mastering the Quantized Intelligence Trade-off via Active Optimization Source: [QuIDE: Mastering the Quantized Intelligence Trade-off via Active Optimization]() Collection index: [llms.txt]() Data updated: 2026-05-14T05:00:05.130Z Authors: Xiantao Jiang Published: 2026-05-05T08:14:28.000Z ## Abstract There is currently no unified metric for evaluating the efficiency of quantized neural networks. We propose QuIDE, built around the Intelligence Index I = (C x P)/log\_2(T+1), which collapses the compression-accuracy-latency trade-off into a single score. Experiments across six settings -- SimpleCNN (MNIST, CIFAR), ResNet-18 (ImageNet-1K), and Llama-3-8B -- show a task-dependent Pareto Knee. 4-bit quantization is optimal for MNIST and large LLMs, while 8-bit is the sweet spot for complex CNN tasks (ResNet-18 on ImageNet), where 4-bit PTQ collapses accuracy catastrophically. The accuracy-gated variant I' correctly flags these non-viable configurations that the raw I would reward. QuIDE provides a reproducible evaluation protocol and a ready-to-use fitness function for mixed-precision search. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Image Classification on CIFAR-10 (Accuracy, Compression, Time, I' Score)](): Image Classification; CIFAR-10; Accuracy: 75.02. - [Image Classification on ImageNet-1K (Accuracy (P), Compression, Time, I' Score)](): Image Classification; ImageNet-1K; Accuracy (P): 69.36. ## Original sources and implementation [Original paper]() [PDF]()