Sunday, October 15, 2023

Researchers find that a modest amount of fine-tuning can undo safety efforts that aim to prevent LLMs such as OpenAI's GPT-3.5 Turbo from spewing toxic content (Thomas Claburn/The Register)

Thomas Claburn / The Register:
Researchers find that a modest amount of fine-tuning can undo safety efforts that aim to prevent LLMs such as OpenAI's GPT-3.5 Turbo from spewing toxic content  —  OpenAI GPT-3.5 Turbo chatbot defenses dissolve with ‘20 cents’ of API tickling  —  The “guardrails” created to prevent large language models …



No comments:

Post a Comment

Russia's finance minister says Russian companies have begun using bitcoin and other digital currencies in international payments to counter Western sanctions (Gleb Bryanski/Reuters)

Gleb Bryanski / Reuters : Russia's finance minister says Russian companies have begun using bitcoin and other digital currencies in i...