SimReg: Achieving Higher Performance in the Pretraining via Embedding Similarity Regularization

📅 2026-05-09
📈 Citations: 0
Influential: 0
📄 PDF
🤖 AI Summary
This work addresses the challenge in large language model pretraining where token embeddings often exhibit high intra-class variance and strong inter-class similarity due to contextual dependencies, thereby limiting representation learning efficiency. For the first time, the study effectively integrates embedding similarity regularization into large-scale language model pretraining by leveraging a contrastive learning mechanism that enhances the clustering of embeddings sharing the same label while enlarging the separation between embeddings of different labels, thus widening the decision margin in multi-class settings. The proposed approach consistently improves both training efficiency and representation quality across dense and mixture-of-experts (MoE) architectures, accelerating convergence by over 30% and boosting average zero-shot performance by more than 1% on standard benchmarks.
📝 Abstract
Pretraining large language models (LLMs) with next-token prediction has led to remarkable advances, yet the context-dependent nature of token embeddings in such models results in high intra-class variance and inter-class similarity, thus hindering the efficiency of representation learning. While similarity-based regularization has demonstrated benefit in supervised fine-tuning and classification tasks, its application and efficacy in large-scale LLM pretraining remains underexplored. In this work, we propose the SimReg, an embedding similarity regularization loss that explicitly encourages token representations with the same ground-truth label within each sequence to be more similar, while enforcing separation from different-label tokens via a contrastive loss. Our analysis reveals that this mechanism introduces gains by enlarging multi-classification margins, thereby enabling more efficient classification. Extensive experiments across dense and Mixture-of-Experts (MoE) architectures demonstrate that SimReg consistently accelerates training convergence by over 30% and improves average zero-shot downstream performance by over 1% across standard benchmarks. Further ablation studies and analyses offer practical insights into hyperparameter tuning and loss effectiveness.
Problem

Research questions and friction points this paper is trying to address.

pretraining
embedding similarity
representation learning
large language models
intra-class variance
Innovation

Methods, ideas, or system contributions that make the work stand out.

embedding similarity regularization
contrastive loss
pretraining efficiency
large language models
representation learning
🔎 Similar Papers
💼 Related Jobs
No related jobs found.
Y
Yan Sun
Baidu Inc.
G
Guoxia Wang
Baidu Inc.
J
Jinle Zeng
Baidu Inc.
J
JiaBin Yang
Baidu Inc.
Shuai Li
Shuai Li
Amazon.com, Inc.
roboticsrobotic manipulationstate estimation
L
Li Shen
Nanyang Technological University
Dacheng Tao
Dacheng Tao
Nanyang Technological University
artificial intelligencemachine learningcomputer visionimage processingdata mining
D
DianHai Yu
Baidu Inc.
Haifeng Wang
Haifeng Wang
Baidu
NLPMTSearchSpeechData Mining