mlsanityData to 14 Oct 2026

QuIDE: Mastering the Quantized Intelligence Trade-off via Active Optimization

arXiv 2605.10959 ↗First posted 5 May 2026cs.LG cs.AIcited by 1

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 14 of its 19 references that are in the map. First posted on arXiv 5 May 2026.

Builds on

14 of 19 references.

References · 14

  1. 2022GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformerscited by 1,919
  2. 2023AWQ: Activation-aware Weight Quantization for LLM Compression and Accelerationcited by 1,298
  3. 2021A Survey of Quantization Methods for Efficient Neural Network Inferencecited by 815
  4. 2015Deep Learning and the Information Bottleneck Principlecited by 802
  5. 2019Learned Step Size Quantizationcited by 628
  6. 2020Up or Down? Adaptive Rounding for Post-Training Quantizationcited by 621
  7. 2021BRECQ: Pushing the Limit of Post-Training Quantization by Block Reconstructioncited by 479
  8. 2018HAQ: Hardware-Aware Automated Quantization with Mixed Precisioncited by 443
  9. 2019HAWQ: Hessian AWare Quantization of Neural Networks with Mixed-Precisioncited by 402
  10. 2020ZeroQ: A Novel Zero Shot Quantization Frameworkcited by 269
  11. 2020Binary Neural Networks: A Surveycited by 190
  12. 2020Integer Quantization for Deep Learning Inference: Principles and Empirical Evaluationcited by 188
2 more
  1. 2021MLPerf Tiny Benchmarkcited by 124
  2. 2020APQ: Joint Search for Network Architecture, Pruning and Quantization Policycited by 99

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026CPR-IE:A Compression-Prediction-Resource Intelligence Efficiency Metriccited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2022QFT: Post-training quantization via fast joint finetuning of all degrees of freedom
  2. 2024Gradient-Free Training of Quantized Neural Networks
  3. 2024QGen: On the Ability to Generalize in Quantization Aware Training
  4. 2023Towards Efficient Verification of Quantized Neural Networks
  5. 2022AMED: Automatic Mixed-Precision Quantization for Edge Devices
  6. 2021Pareto-Optimal Quantized ResNet Is Mostly 4-bit
  7. 2025Batch Normalization-Free Fully Integer Quantized Neural Networks via Progressive Tandem Learning
  8. 2025InfoQ: Mixed-Precision Quantization via Global Information Flow