English

DLR: Zero-Inference-Cost Latent Residuals for Low-Rank Pre-Training

Machine Learning 2026-06-27 v1 Artificial Intelligence

Abstract

Large language models have driven recent progress in language and multimodal AI, yet pre-training them at scale is prohibitively expensive. Low-rank pre-training, which factorizes each weight matrix into a rank-r product to reduce both parameters and FLOPs, is a promising response but typically lags full-rank training in quality. We propose Duplicated Latent Residual (DLR), a training-only, parameter-free, foldable plug-in for low-rank pre-training. DLR augments the standard low-rank output Bz with a fixed structured residual alpha/sqrt(K) * Expand_K(z) that replicates each latent coordinate K = ceil(d_out/r) times across the output. With alpha fixed, DLR adds zero learnable parameters per layer; after training, it is absorbed into the up-projection in closed form, B* = B + alpha/sqrt(K) R^T, so deployment parameter count, FLOPs and memory match the underlying low-rank backbone exactly. Across LLaMA models from 60M to 7B parameters, DLR strengthens low-rank pre-training on C4 validation perplexity in most settings, with the clearest gains at 130M and above; folded checkpoints transfer cleanly to supervised fine-tuning on standard benchmarks.

Keywords

Cite

@article{arxiv.2606.28932,
  title  = {DLR: Zero-Inference-Cost Latent Residuals for Low-Rank Pre-Training},
  author = {Dong Wang and Wenwu Tang and Yun Cheng and Olga Saukh},
  journal= {arXiv preprint arXiv:2606.28932},
  year   = {2026}
}

Comments

Includes appendix, 6 figures and 11 tables. Code available at https://github.com/nanguoyu/DLR