Paul Janson

12 papers A* 2Journal 9Unranked 1
YearRankTypeTitle / Venue / Authors
2026 J jnl
Heterogeneous Low-Bandwidth Pre-Training of LLMs.
CoRR
Yazan Obeidi, Amir Sarfi, Joel Lidin, Paul Janson, Eugene Belilovsky
2026 J jnl
Stabilizing Native Low-Rank LLM Pretraining.
CoRR
Paul Janson, Edouard Oyallon, Eugene Belilovsky
2025 J jnl
Beyond Cosine Decay: On the effectiveness of Infinite Learning Rate Schedule for Continual Pre-training.
CoRR
Vaibhav Singh, Paul Janson, Paria Mehrbod, Adam Ibrahim, Irina Rish, Eugene Belilovsky, Benjamin Thérien
2025 J jnl
PyLO: Towards Accessible Learned Optimizers in PyTorch.
CoRR
Paul Janson, Benjamin Thérien, Quentin Anthony, Xiaolong Huang, Abhinav Moudgil, Eugene Belilovsky
2024 J jnl
FewShotNeRF: Meta-Learning-based Novel View Synthesis for Rapid Scene-Specific Adaptation.
CoRR
Piraveen Sivakumar, Paul Janson, Jathushan Rajasegaran, Thanuja D. Ambegoda
2024 A* conf
Overcoming Generic Knowledge Loss with Selective Parameter Update.
CVPR
Wenxuan Zhang, Paul Janson, Rahaf Aljundi, Mohamed Elhoseiny
2024 — conf
Towards Motion from Video Diffusion Models.
ECCV Workshops (12)
Paul Janson, Tiberiu Popa, Eugene Belilovsky
2024 J jnl
Towards motion from video diffusion models.
CoRR
Paul Janson, Tiberiu Popa, Eugene Belilovsky
2023 A* conf
Continual Zero-Shot Learning through Semantically Guided Generative Random Walks.
ICCV
Wenxuan Zhang, Paul Janson, Kai Yi, Ivan Skorokhodov, Mohamed Elhoseiny
2023 J jnl
Continual Zero-Shot Learning through Semantically Guided Generative Random Walks.
CoRR
Wenxuan Zhang, Paul Janson, Kai Yi, Ivan Skorokhodov, Mohamed Elhoseiny
2023 J jnl
Overcoming General Knowledge Loss with Selective Parameter Finetuning.
CoRR
Wenxuan Zhang, Paul Janson, Rahaf Aljundi, Mohamed Elhoseiny
2022 J jnl
A Simple Baseline that Questions the Use of Pretrained-Models in Continual Learning.
CoRR
Paul Janson, Wenxuan Zhang, Rahaf Aljundi, Mohamed Elhoseiny