9 papers
Alignment Reduces Expressed but Not Encoded Gender Bias: A Unified Framework and Study
Nour Bouchouchi, Thibault Laugel, Xavier Renard +3
During training, Large Language Models (LLMs) learn social regularities that can lead to gender bias in downstream applications. Most mitigation efforts focus on reducing bias in g…
From Demographics to Survey Anchors: Evaluating LLM Agents for Modeling Retirement Attitudes
Rubén Garzón, Pauline Baron, Vincent Grari +3
Large language models (LLM) agents may offer tools to predict human responses to surveys. A common technique for defining these agents uses only demographics, for example country,…
ACT: Agentic Classification Tree
Vincent Grari, Tim Arni, Thibault Laugel +3
When used in high-stakes settings, AI systems are expected to produce decisions that are transparent, interpretable and auditable, a requirement increasingly expected by regulation…
Agentic Adversarial QA for Improving Domain-Specific LLMs
Vincent Grari, Ciprian Tomoiaga, Sylvain Lamprier +2
Large Language Models (LLMs), despite extensive pretraining on broad internet corpora, often struggle to adapt effectively to specialized domains. There is growing interest in fine…
SAKE: Steering Activations for Knowledge Editing
Marco Scialanga, Thibault Laugel, Vincent Grari +1
As Large Langue Models have been shown to memorize real-world facts, the need to update this knowledge in a controlled and efficient manner arises. Designed with these constraints…
Metric assessment protocol in the context of answer fluctuation on MCQ tasks
Ekaterina Goliakova, Xavier Renard, Marie-Jeanne Lesot +3
Using multiple-choice questions (MCQs) has become a standard for assessing LLM capabilities efficiently. A variety of metrics can be employed for this task. However, previous resea…