Zero-hallucination knowledge engine – LLM never reasons, graph does all the work(github.com)
github.com
Zero-hallucination knowledge engine – LLM never reasons, graph does all the work
https://github.com/skvcool-rgb/KOS-Engine
5 comments
Claims of "Zero-hallucination" usually fall apart when the engine has to derive a new insight from the graph. If the LLM isn't reasoning, how are you enforcing Fidelity between the graph source and the final output?
What's the overhead in terms of time and whats the actual mechanisms for provability? Additionally, what breaks because of this, surely there are downsides to Extreme Typo Phonetics correction.
Share your feedback