Roman Plaud — accepted AI-conference papers

2026

Tailoring Strictly Proper Scoring Rules for Downstream Tasks: An Application to Causal Inference

Roman Plaud, Alexandre Perez-Lebel, Antoine Saillenfest, Thomas Bonald, Marine Le Morvan, Gael Varoquaux +1

ICML 2026poster

Probabilistic models are typically trained using task-agnostic objectives like log-loss, which can lead to significant errors in downstream estimation. This disconnect is especially critical in Inverse Probability Weighting (IPW) for causal inference, where propensity score errors near $0$ and $1$ o…

Cited by 1 Source ↗Scholar ↗

2025

Ignore the KL Penalty! Boosting Exploration on Critical Tokens to Enhance RL Fine-Tuning

Jean Vassoyan, Nathanaël Beau, Roman Plaud

NAACL 2025findings

The ability to achieve long-term goals is a key challenge in the current development of large language models (LLMs). To address this, pre-trained LLMs can be fine-tuned with reinforcement learning (RL) to explore solutions that optimize a given goal. However, exploration with LLMs is difficult, as…

Cited by 2 Source ↗PDF ↗code ↗Scholar ↗

2025

To Each Metric Its Decoding: Post-Hoc Optimal Decision Rules of Probabilistic Hierarchical Classifiers

Roman Plaud, Alexandre Perez-Lebel, Matthieu Labeau, Antoine Saillenfest, Thomas Bonald

ICML 2025poster

Hierarchical classification offers an approach to incorporate the concept of mistake severity by leveraging a structured, labeled hierarchy. However, decoding in such settings frequently relies on heuristic decision rules, which may not align with task-specific evaluation metrics. In this work, we p…

Cited by 0 Source ↗PDF ↗Scholar ↗