Skip to content
AI News HubLIVE
Original source2 min read

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

Summary

Researchers propose QuIDE, a unified metric that combines compression, accuracy, and latency into a single Intelligence Index score for evaluating quantized neural networks. Experiments across various architectures reveal task-dependent optimal quantization levels, with 4-bit working best for simple tasks and large language models, while 8-bit is optimal for complex image classification. An accuracy-gated variant prevents overvaluing configurations that destroy accuracy.

SourcearXiv Machine LearningAuthor: Xiantao Jiang
QuIDE: Mastering the Quantized Intelligence Trade-off via Active Optimization
Report an error

The correction channel is not available yet. You can copy the article reference below for later.

Correction instructions
Read article

[2605.10959] QuIDE: Mastering the Quantized Intelligence Trade-off via Active Optimization

[Submitted on 5 May 2026]

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

View a PDF of the paper titled QuIDE: Mastering the Quantized Intelligence Trade-off via Active Optimization, by Xiantao Jiang

View PDF HTML (experimental)

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.

Comments: 16 pages, 9 figures

Subjects:

Machine Learning (cs.LG); Artificial Intelligence (cs.AI)

ACM classes: I.2.6; I.5.1

Cite as: arXiv:2605.10959 [cs.LG]

(or arXiv:2605.10959v1 [cs.LG] for this version)

https://doi.org/10.48550/arXiv.2605.10959

arXiv-issued DOI via DataCite

Submission history

From: Xiantao Jiang [view email] [v1] Tue, 5 May 2026 08:14:28 UTC (1,418 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled QuIDE: Mastering the Quantized Intelligence Trade-off via Active Optimization, by Xiantao Jiang

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.LG

new | recent | 2026-05

Change to browse by:

cs cs.AI

References & Citations

NASA ADS

Google Scholar

Semantic Scholar

Loading...

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

IArxiv recommender toggle

IArxiv Recommender (What is IArxiv?)

Author

Venue

Institution

Topic

About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)

Key points and analysis

Article intelligence

InvestorsAdvanced

Key points

  • QuIDE introduces the Intelligence Index I = (C × P) / log2(T+1) to collapse the compression-accuracy-latency trade-off into a single score.
  • Experiments show that 4-bit quantization is optimal for MNIST and large LLMs like Llama-3-8B, while 8-bit is the sweet spot for complex CNNs like ResNet-18 on ImageNet.
  • The accuracy-gated variant I' correctly identifies non-viable configurations where 4-bit quantization causes catastrophic accuracy loss.
  • QuIDE provides a reproducible evaluation protocol and a fitness function for mixed-precision search.

Highlights and analysis are generated automatically and may contain errors. Check the original source.

More recent stories on this topic · Models

QuIDE: Mastering the Quantized Intelligence Trade-off via Active Optimization | AI News Hub