← Search

Inés García-Redondo

2 accepted papers

2026

The Shape of Adversarial Influence: Characterizing LLM Latent Spaces with Persistent Homology

ICLR 2026oral

Existing interpretability methods for Large Language Models (LLMs) often fall short by focusing on linear directions or isolated features, overlooking the high-dimensional, nonlinear, and relational geometry within model representations. This study focuses on how adversarial inputs systematically af…

Cited by 0SourceScholar
2024

On the Limitations of Fractal Dimension as a Measure of Generalization

NeurIPS 2024poster

Bounding and predicting the generalization gap of overparameterized neural networks remains a central open problem in theoretical machine learning. There is a recent and growing body of literature that proposes the framework of fractals to model optimization trajectories of neural networks, motivati…