OptimalBrainCompression: AFrameworkfor AccuratePost-Training QuantizationandPruning

Feb-7-2026, 18:56:10 GMT–Neural Information Processing Systems

Our frameworkstarts from thelayer-wise compression problem described above,bywhich theglobal compression task,defined either forpruning orquantization, is first split into layer-wise sub-problems, based on the layer behavior on the calibration data.

artificial intelligence, machine learning, quantization, (13 more...)

Neural Information Processing Systems

Feb-7-2026, 18:56:10 GMT

Conferences PDF

Add feedback

Technology:
- Information Technology > Artificial Intelligence
  - Machine Learning (0.95)
  - Representation & Reasoning (0.68)

Duplicate Docs Excel Report

Title
Optimal Brain Compression: A Framework for Accurate Post-Training Quantization and Pruning

Similar Docs Excel Report more

Title	Similarity	Source
None found