steadeepanda

6 ms·

1. 1.▲
   by steadeepanda 6mo ago

   Here is the GitHub link: https://github.com/steadeepanda/agent-ruler/
2. 2.▲
   Show HN: Agent Ruler new update v0.1.9

   2 points by steadeepanda 6mo ago|1 comments
3. 3.▲
   by steadeepanda 6mo ago

   Hmmm yes and no, here the output "tends towards" deterministic (≠ deterministic), because the internal process contains noise (noise can't be nullified, otherwise LLMs wouldn't make mistakes and wouldn't produce une
4. 4.▲
   by steadeepanda 6mo ago

   Ah okayy, yeah sure you're right. I didn't mean it that way. I mean I know we can get deterministic output from LLM but the issue is that even with that LLMs are trained on large set of data that open a surface for prompt injectio
5. 5.▲
   by steadeepanda 6mo ago

   LLMs are probabilistic by nature so even if you're using without temperature it doesn't remove completely this fact, it would just narrow the output. However here we're aiming for an already defined set of rules on purpose,
6. 6.▲
   by steadeepanda 6mo ago

   Thanks
7. 7.▲
   Show HN: Deterministic security solution for AI agents – OpenClaw and 2 more

   4 points by steadeepanda 6mo ago|8 comments