GitHub repository for DORA: Data-agnOstic Representation Analysis paper. DORA allows to find outlier representations in Deep Neural Networks.

Interpretability study of LoRA fine-tuning in LLMs: where LoRA puts its capacity (per-module update norm + effective rank) and how much it shifts representations (layer-wise drift), with rank/module ablations. Qwen2.5-1.5B, paper + code.

Attribution-based analysis of French grammatical gender encoding in FlauBERT embeddings using SHAP, LIME, Random Forest, and cross-validation methods across multiple architectures

Applied research in representation and decision behavior

🌐 Investigate and understand coherence in complex information environments to enhance structural clarity and decision-making across varying scales.