AI Tools & Models HOTFeaturedAll
arXiv cs.AIScore75

ROCKET: Rapid Optimization via Calibration-guided Knapsack Enhanced Truncation for Efficient Model Compression

Summary

ROCKET is a training-free model compression method that uses calibration-guided truncation to achieve leading performance at high compression ratios.

Why this matters

Covers model and API · Relevant to this topic

Open original ↗

Source: arXiv cs.AI · arxiv.org

Topics

Related items