AI & ML
impact 16
The Shape of Adversarial Influence: Characterizing LLM Latent Spaces with Persistent Homology
The Shape of Adversarial Influence: Characterizing LLM Latent Spaces with Persistent Homology arXiv:2505.20435v3 Announce Type: replace-cross Abstract: Existing interpretability methods for Large Language Models (LLMs)…
Why it matters
A useful signal for anyone monitoring adversarial. The shape factor makes this more consequential than it first appears.