Meta Paper: Feature-Importance Embedding Pruning Cuts Recommendation Model Memory 20-40%

_reachsumit · x · 2026-11-20

Meta researchers published FI-ITEP, an in-training embedding pruning method that uses feature importance (instead of size-based heuristics) to set per-table pruning ratios in large recommendation models.

Original post →

More from Infra

Infra channel →