DimGrow: Memory-Efficient Field-level Embedding Dimension Search

arXiv:2505.12683 · cs.LG · Submitted 2025-05-19 · Read on arXiv

cs.LG

Submitted: 2025-05-19

Updated: 2026-09-29

Terminology

• Pruning — 1× in this paper · explained in 4 episodes

Sources

• Post-Training 4-bit Quantization on Embedding Tables
• DARTS: Differentiable Architecture Search
• Learnable Embedding Sizes for Recommender Systems
• Field-wise Embedding Size Search via Structural Hard Auxiliary Mask Pruning for Click-Through Rate Prediction
• Mixed-Precision Embedding Using a Cache
• Serving Deep Learning Models with Deduplication from Relational Databases

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks