3 papers
cs.CY2026
StereoTales: A Multilingual Framework for Open-Ended Stereotype Discovery in LLMs
Pierre Le Jeune, Ãtienne Duchesne, Weixuan Xiao +4
Multilingual studies of social bias in open-ended LLM generation remain limited: most existing benchmarks are English-centric, template-based, or restricted to recognizing pre-spec…
math.OC2025
Analysis and Synthesis Denoisers for Forward-Backward Plug-and-Play Algorithms
Matthieu Kowalski, Benoît Malézieux, Thomas Moreau +1
In this work we study the behavior of the forward-backward (FB) algorithm when the proximity operator is replaced by a sub-iterative procedure to approximate a Gaussian denoiser, i…
cs.CY2025
Phare: A Safety Probe for Large Language Models
Pierre Le Jeune, Benoît Malézieux, Weixuan Xiao +1
Ensuring the safety of large language models (LLMs) is critical for responsible deployment, yet existing evaluations often prioritize performance over identifying failure modes. We…