List your company

Unlearning Evaluation on Harry Potter Q&A

30.9Embedding Space ScoreEmbedding Space Attack
19.56422.50725.4528.393Feb 14, 2024

Evaluation results

MethodLinks
Embedding Space AttackAttack Type=Universal, Suffix Tokens=1, Backbone=LlamaHP2024.0230.9———
Embedding Space AttackAttack Type=Universal, Suffix Tokens=5, Backbone=LlamaHP2024.0230.9———
Embedding Space AttackAttack Type=Individual, Suffix Tokens=1, Backbone=LlamaHP2024.0225.5———
Embedding Space AttackAttack Type=Universal, Suffix Tokens=20, Backbone=LlamaHP2024.0225.5———
Embedding Space AttackAttack Type=Individual, Suffix Tokens=5, Backbone=LlamaHP2024.0221.8———
Embedding Space AttackAttack Type=Individual, Suffix Tokens=20, Backbone=LlamaHP2024.0220———
Head Projection Attack (HPA)Attack Type=Individual, Backbone=LlamaHP2024.02——7.2—
Probability Delta Attack (PDA)Attack Type=Individual, Backbone=LlamaHP2024.02———9