LLMs That Lie Logically: New Backdoor Turns Reasoning Into a Weapon
A new LLM backdoor called ‘alibi-aligned reasoning’ hides malicious behavior inside logical inference, making it far harder to detect than trigger-word attacks.
Backdoor
A new LLM backdoor called ‘alibi-aligned reasoning’ hides malicious behavior inside logical inference, making it far harder to detect than trigger-word attacks.