Curriculum Vitae
Download PDF
Ph.D. Candidate in Computer Science, National University of Singapore · Jan 2023 – Jan 2027
yannis.montreuil@u.nus.edu ·
Google Scholar ·
GitHub ·
LinkedIn
Research Profile
I develop statistical learning theory and algorithms for reliable, cost-aware AI systems that decide when to answer directly and when to delegate to stronger models, tools, or experts. My research studies learning-to-defer, model routing, and multi-expert allocation, with guarantees under accuracy, cost, robustness, online adaptation, and distribution-shift constraints.
I expect to complete my PhD in January 2027. I am looking for research internships in quantitative research and machine learning research, and I am open to full-time roles.
Publications
Peer-reviewed
- Why Ask One When You Can Ask k? Learning-to-Defer to the Top-k Experts. Yannis Montreuil, Axel Carlier, Lai Xing Ng, Wei Tsang Ooi. ICLR 2026. arXiv:2504.12988
- Online Learning-to-Defer with Varying Experts. Yannis Montreuil*, Dang Hoang Duy*, Maxime Meyer*, Axel Carlier, Lai Xing Ng, Wei Tsang Ooi. AISTATS 2026. arXiv:2605.12340
- Adversarial Robustness in One-Stage Learning-to-Defer. Yannis Montreuil*, Letian Yu*, Axel Carlier, Lai Xing Ng, Wei Tsang Ooi. AISTATS 2026. arXiv:2510.10988
- Optimal Query Allocation in Extractive QA with LLMs: A Learning-to-Defer Framework with Theoretical Guarantees. Yannis Montreuil*, Yeo Shu Heng*, Axel Carlier, Lai Xing Ng, Wei Tsang Ooi. AISTATS 2026. arXiv:2410.15761
- Adversarial Robustness in Two-Stage Learning-to-Defer: Algorithms and Guarantees. Yannis Montreuil, Axel Carlier, Lai Xing Ng, Wei Tsang Ooi. ICML 2025. arXiv:2502.01027
- A Two-Stage Learning-to-Defer Approach for Multi-Task Learning. Yannis Montreuil*, Yeo Shu Heng*, Axel Carlier, Lai Xing Ng, Wei Tsang Ooi. ICML 2025. arXiv:2410.15729
- Towards Robust Human–AI Decision-Making via Learning-to-Defer. Yannis Montreuil. AAAI-26 Doctoral Consortium.
Preprints
- Beyond Augmented-Action Surrogates for Multi-Expert Learning-to-Defer. Yannis Montreuil, Axel Carlier, Lai Xing Ng, Wei Tsang Ooi. arXiv:2604.09414
- Learning-to-Defer with Expert-Conditioned Advice. Yannis Montreuil, Leina Montreuil, Axel Carlier, Lai Xing Ng, Wei Tsang Ooi. arXiv:2603.14324
- Learning to Defer in Non-Stationary Time Series via Switching State-Space Models. Yannis Montreuil*, Letian Yu*, Axel Carlier, Lai Xing Ng, Wei Tsang Ooi. arXiv:2601.22538
* indicates equal contribution.
Research Experience
- First-author research on learning-to-defer, model routing, expert allocation, online learning, and robustness.
- Algorithms and guarantees for deciding when lightweight predictors should act directly or delegate to stronger experts or models.
- Analysis of deferral policies under cost constraints, adversarial perturbations, non-stationarity, and changing expert availability.
- PyTorch pipelines for deferral algorithms, LLM evaluation, and distributed experiments.
- Theoretical frameworks for routing and deferral in decision-making systems under uncertainty.
- Allocation and delegation policies with guarantees for hybrid AI systems and smart-city decision pipelines.
- Collaborated with Prof. Wei Tsang Ooi, Prof. Axel Carlier, and Dr. Lai Xing Ng.
- Deep-learning and geometric methods for 6-DoF robotic manipulation of deformable objects.
- Self-supervised and implicit-learning pipelines for 3D point-cloud data.
- Improved building detection in satellite imagery using semi-supervised learning, including Mean Teacher and Naive Student.
- Increased detection accuracy by 30% through data augmentation, architectural refinements, and hyperparameter tuning.
Education
A*STAR Merit Award; AISG–DesCartes Joint Excellence Scholarship. Research focus: statistical learning theory, learning-to-defer, model routing, online learning, robustness, and multi-expert systems.
Awards
Academic Service
Reviewer: ICML 2026 (Silver Reviewer), ICLR 2026, AISTATS 2026.
Teaching: CS5242 Neural Networks and Deep Learning; CP2107 Independent Introduction to CS Research.
Students supervised: Yeo Shu Heng, Dang Hoang Duy, Letian Yu, Seah Min-Rei.
Technical Skills
| Theory | Statistical learning theory, online learning, robustness, optimization, probability, inference. |
|---|---|
| Research areas | Learning-to-defer, selective prediction, model routing, expert allocation, multi-expert systems, decision-making under uncertainty. |
| Methods | Formal proofs, algorithm design, regret and generalization analysis, experimental evaluation, scientific writing. |
| ML engineering | PyTorch, distributed training, LLM evaluation, large-scale ML experimentation, supervised learning, semi-supervised learning, deep learning. |
| Optimization tools | MOSEK, convex optimization, numerical optimization, Python scientific stack, C, R. |
| Languages | French (native), English (fluent), Spanish (intermediate). |
| Interests | Muay Thai, tennis, table tennis, League of Legends when I find the time. |