1 citations · 2 across the 7 of their papers we have counts for
10 papers
Kimina-Prover Preview: Towards Large Formal Reasoning Models with Reinforcement Learning
Haiming Wang, Mert Unsal, Xiaohan Lin +37
We introduce Kimina-Prover Preview, a large language model that pioneers a novel reasoning-driven exploration paradigm for formal theorem proving, as showcased in this preview rele…
Transferability Metrics for Object Detection
Louis Fouquet, Simona Maggio, Léo Dreyfus-Schmidt
Transfer learning aims to make the most of existing pre-trained models to achieve better performance on a new task in limited data scenarios. However, it is unclear which models wi…
Towards Clear Expectations for Uncertainty Estimation
Victor Bouvier, Simona Maggio, Alexandre Abraham +1
If Uncertainty Quantification (UQ) is crucial to achieve trustworthy Machine Learning (ML), most UQ methods suffer from disparate and inconsistent evaluation protocols. We claim th…
Performance Prediction Under Dataset Shift
Simona Maggio, Victor Bouvier, Léo Dreyfus-Schmidt
ML models deployed in production often have to face unknown domain changes, fundamentally different from their training settings. Performance prediction models carry out the crucia…
Sample Noise Impact on Active Learning
Alexandre Abraham, Léo Dreyfus-Schmidt
This work explores the effect of noisy sample selection in active learning strategies. We show on both synthetic problems and real-life use-cases that knowledge of the sample noise…
Ensembling Shift Detectors: an Extensive Empirical Evaluation
Simona Maggio, Léo Dreyfus-Schmidt
The term dataset shift refers to the situation where the data used to train a machine learning model is different from where the model operates. While several types of shifts natur…
Rebuilding Trust in Active Learning with Actionable Metrics
Alexandre Abraham, Léo Dreyfus-Schmidt
Active Learning (AL) is an active domain of research, but is seldom used in the industry despite the pressing needs. This is in part due to a misalignment of objectives, while rese…
Dataiku's Solution to SPHERE's Activity Recognition Challenge
Maxime Voisin, Leo Dreyfus-Schmidt, Pierre Gutierrez +2
Our team won the second prize of the Safe Aging with SPHERE Challenge organized by SPHERE, in conjunction with ECML-PKDD and Driven Data. The goal of the competition was to recogni…
Stability conditions on Brauer tree algebras
Léo Dreyfus-Schmidt
We study the space of stability conditions attached to the derived category of -mod for the Brauer tree algebra of the line with edges. These algebras arise in t…
Splendid and perverse equivalences
Léo Dreyfus-Schmidt
Inspired by the works of Rickard on splendid equivalences and of Chuang and Rouquier on perverse equivalences, we are here interested in the combination of both, a splendid pervers…