Talks / Posters
A Post-Processing Conformal Prediction Approach for Conditional Coverage via Pivotal Scores
Poster · Ukrainian Mathematics Days, Institut Henri Poincaré, June 2026
Discusses the PIT-CP framework for distribution-free conditional validity in conformal prediction.
A General Framework for Joint Multi-State Models
Talk · Journées de la Statistique (SFdS), Clermont-Ferrand, June 2026
Introduces a unified framework for jointly modeling longitudinal biomarkers and multi-state event processes on directed graphs.
Uniform High-Probability Bands for Survival Functions
Poster · StatMathAppli, Fréjus, September 2025
Presents asymptotic uniform confidence bands for survival estimators derived under a joint modelling framework. The bands hold with high probability uniformly over time, providing rigorous inferential guarantees for censored survival data beyond classical pointwise confidence intervals.
Selecting the Number of Clusters in a Clustering Algorithm
Talk · Journées de la Statistique (SFdS), Marseille, June 2025
Introduces a data-driven model selection criterion for choosing the number of classes in the Spectral Bridges clustering algorithm. The method leverages the scale-invariant affinity structure of the algorithm to define a connectivity-based score, avoiding cross-validation and working directly from the spectral decomposition.
Service & Teaching
| Date | Activity |
|---|---|
| October 2026 | Reviewer for AISTATS 2027 (upcoming) |
| September 2026 | Organizer of LLM sessions at LaMME |
| May 2026 | Co-organizer of a Causality Working Group at LaMME with Pierre Humbert |
| 2026–2027 | Teaching assistant (TDs), Université d’Évry, 60 hours/year expected |
Attended Events
| Date | Event |
|---|---|
| March 2026 | Workshop on Uncertainty Quantification and Causality (CAUSALI-T-AI, Jussieu) |
| October 2025 | Winter School on Causality and Explainable AI (SCAI, Jussieu) |
| July 2025 | Working group on Topological Data Analysis (LaMME) |
| June 2025 | Working group on Causality (MIA Paris-Saclay) |