Scaling Laws for Adversarial Attacks on Language Model Activations and Tokens

Stanislav Fort. Scaling Laws for Adversarial Attacks on Language Model Activations and Tokens. In The Thirteenth International Conference on Learning Representations, ICLR 2025, Singapore, April 24-28, 2025. OpenReview.net, 2025. [doi]

Authors

Stanislav Fort

This author has not been identified. Look up 'Stanislav Fort' in Google