Oleksiy Ostapenko
Research Scientist
Model Readiness
I am a Senior Research Scientist at ServiceNow AI Research, where I worked on building Apriel family of models and specifically it’s efficient reasoning variants, including Super Apriel. I am currently working on agentic post-training of LLMs for enterprise applications. My interests span efficient deep learning, continual learning, modularity and long-horizon RL. Previously, I completed my Ph.D at Mila - Quebec Artificial Intelligence Institute under the supervision of Professor Laurent Charlin. During this time I worked as visiting researcher at Microsoft Research and ServiceNow AI Research. Before that I worked on continual learning at SAP Research with Moin Nabi and Tassilo Klein.
Interests
- Continual Learning
Publications
DiffuMamba: High-Throughput Diffusion LMs with Mamba Backbone.
International Conference on Machine Learning (ICML),
2026.
Using Scaling Laws for Data Source Utility Estimation in Domain-Specific Pre-Training.
Conference on Language Modeling Workshops,
2025.
Exploring Sparse Adapters for Scalable Merging of Parameter Efficient Experts.
Conference on Language Modeling (COLM),
2025.
Attention for Compositional Modularity.
Workshop at the Neural Information Processing Systems (NeurIPS),
2022.
Continual Learning with self-selecting specialized modules through expansion and pruning.
Montreal AI Symposium (MAIS),
2022.
Continual Learning with Foundation Models: An Empirical Study of Latent Replay.
Conference on Lifelong Learning Agents (CoLLAs),
2022.
Continual Learning with Foundation Models: An Empirical Study of Latent Replay.
Workshop at the Conference on Computer Vision and Pattern Recognition (CVPR),
2022.
Continual Learning via Local Module Composition.
Conference on Neural Information Processing Systems (NeurIPS),
2021.
Online Fast Adaptation and Knowledge Accumulation: a New Approach to Continual Learning.
Conference on Neural Information Processing Systems (NeurIPS),
2020.
Online Fast Adaptation and Knowledge Accumulation: a New Approach to Continual Learning.
Workshop at the Conference on Computer Vision and Pattern Recognition (CVPR),
2020.