Large Language Model Unlearning via Embedding-Corrupted Prompts
–Neural Information Processing Systems
Instead of relying on the LLM itself to unlearn, we enforce an unlearned state during inference by employing a prompt classifier to identify and safeguard prompts to forget.
Neural Information Processing Systems
Oct-10-2025, 17:55:43 GMT
- Country:
- Asia
- China (0.04)
- India > Maharashtra
- Mumbai (0.04)
- Kazakhstan (0.04)
- Middle East > Kuwait
- Capital Governorate > Kuwait City (0.04)
- Europe
- Central Europe (0.04)
- Italy > Calabria
- Catanzaro Province > Catanzaro (0.04)
- Montenegro (0.04)
- Poland (0.04)
- North America > United States
- California
- San Diego County > San Diego (0.04)
- Santa Cruz County > Santa Cruz (0.04)
- Massachusetts (0.04)
- Virginia (0.04)
- California
- South America > Chile
- Asia
- Genre:
- Research Report > Experimental Study (1.00)
- Industry:
- Education (1.00)
- Government (1.00)
- Health & Medicine > Therapeutic Area
- Psychiatry/Psychology (0.45)
- Information Technology > Security & Privacy (1.00)
- Law (1.00)
- Leisure & Entertainment (1.00)
- Media (1.00)
- Technology: