Dual-objective Language Models: Training Efficiency Without Overfitting

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Samuel, David, Charpentier, Lucas Georges Gabriel
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918412202016768
author Samuel, David
Charpentier, Lucas Georges Gabriel
author_facet Samuel, David
Charpentier, Lucas Georges Gabriel
contents This paper combines autoregressive and masked-diffusion training objectives without any architectural modifications, resulting in flexible language models that outperform single-objective models. Autoregressive modeling has been a popular approach, partly because of its training efficiency; however, that comes at the cost of sensitivity to overfitting. On the other hand, masked-diffusion models are less efficient to train while being more resilient to overfitting. In this work, we demonstrate that dual-objective training achieves the best of both worlds. To derive the optimal balance between both objectives, we train and evaluate 50 language models under varying levels of data repetition. We show that it is optimal to combine both objectives under all evaluated settings and that the optimal balance is similar whether targeting autoregressive or masked-diffusion downstream performance.
format Preprint
id arxiv_https___arxiv_org_abs_2512_14549
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Dual-objective Language Models: Training Efficiency Without Overfitting
Samuel, David
Charpentier, Lucas Georges Gabriel
Computation and Language
Artificial Intelligence
This paper combines autoregressive and masked-diffusion training objectives without any architectural modifications, resulting in flexible language models that outperform single-objective models. Autoregressive modeling has been a popular approach, partly because of its training efficiency; however, that comes at the cost of sensitivity to overfitting. On the other hand, masked-diffusion models are less efficient to train while being more resilient to overfitting. In this work, we demonstrate that dual-objective training achieves the best of both worlds. To derive the optimal balance between both objectives, we train and evaluate 50 language models under varying levels of data repetition. We show that it is optimal to combine both objectives under all evaluated settings and that the optimal balance is similar whether targeting autoregressive or masked-diffusion downstream performance.
title Dual-objective Language Models: Training Efficiency Without Overfitting
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2512.14549