6 papers
Shieldstral
Antonia Calvi, Avinash Sooriyarachchi, Giada Pistilli +274
We introduce Shieldstral, a 3B-parameter policy-adaptive multimodal safety classifier that matches or outperforms models nearly 7 its size on text safety benchmarks and set…
Multiple Choice Learning of Low-Rank Adapters for Language Modeling
Victor Letzelter, Hugo Malard, Mathieu Fontaine +4
We propose LoRA-MCL, a training scheme that extends next-token prediction in language models with a method designed to decode diverse, plausible sentence continuations at inference…
Test-Time Conditioning with Representation-Aligned Visual Features
Nicolas Sereyjol-Garros, Ellington Kirby, Victor Letzelter +2
While representation alignment with self-supervised models has been shown to improve diffusion model training, its potential for enhancing inference-time conditioning remains large…
Winner-takes-all for Multivariate Probabilistic Time Series Forecasting
Adrien Cortés, Rémi Rehm, Victor Letzelter
We introduce TimeMCL, a method leveraging the Multiple Choice Learning (MCL) paradigm to forecast multiple plausible time series futures. Our approach employs a neural network with…
Annealed Winner-Takes-All for Motion Forecasting
Yihong Xu, Victor Letzelter, Mickaël Chen +2
In autonomous driving, motion prediction aims at forecasting the future trajectories of nearby agents, helping the ego vehicle to anticipate behaviors and drive safely. A key chall…
Annealed Multiple Choice Learning: Overcoming limitations of Winner-takes-all with annealing
David Perera, Victor Letzelter, Théo Mariotte +4
We introduce Annealed Multiple Choice Learning (aMCL) which combines simulated annealing with MCL. MCL is a learning framework handling ambiguous tasks by predicting a small set of…