News for hackers. Stuff that matters.
Re:AI is becoming 'yes-men on servers' by janalsncm on Monday March 10, 2025 @04:47PM (#43325895)
One approach does use this. You can ask an LLM to explicitly check its own answers by outputting thinking tokens, generating a reward signal if it gets the right answer, and directly updating based on the reward signals. That’s a part of how DeepSeek R1 was trained. It’s better but not perfect, because the thinking process is imperfect. Ultimately the LLM might not know what it doesn’t know.

There are no comments on this comment yet.