collaborators

5 papers

cs.LG2026

Stabilizing Native Low-Rank LLM Pretraining

Paul Janson, Edouard Oyallon, Eugene Belilovsky

The paper presents a method to train large language models from scratch using only low‑rank factorized weights by introducing Spectron, a spectral renormalization and orthogonaliza…

cs.LG2026

Learned Subspace Compression for Communication-Efficient Pipeline Parallelism

Paul Janson, Edouard Oyallon, Eugene Belilovsky

Pipeline parallelism enables training of large language models that exceed single-device memory, yet inter-stage activation communication becomes the dominant bottleneck when train…

cs.LG2026

PyLO: Towards Accessible Learned Optimizers in PyTorch

Paul Janson, Benjamin Therien, Quentin Anthony +3

Learned optimizers have been an active research topic over the past decade, with increasing progress toward practical, general-purpose optimizers that can serve as drop-in replacem…

cs.LG2026

Heterogeneous Low-Bandwidth Pre-Training of LLMs

Yazan Obeidi, Amir Sarfi, Joel Lidin +2

Pre-training large language models (LLMs) increasingly requires distributed compute, yet bandwidth constraints make it difficult to scale beyond well-provisioned datacenters-especi…

cs.LG2025

Beyond Cosine Decay: On the effectiveness of Infinite Learning Rate Schedule for Continual Pre-training

Vaibhav Singh, Paul Janson, Paria Mehrbod +4

The ever-growing availability of unlabeled data presents both opportunities and challenges for training artificial intelligence systems. While self-supervised learning (SSL) has em…