Суть, в общем-то, в выводах:
In this paper, we study the fundamental problem of eliminating hallucinations in LLMs. To do so, we define a formal world where hallucination in LLMs can be clearly defined and discussed. Specifically, hallucination is defined as inconsistencies between computable LLMs and a computable ground truth function. By utilizing results in learning theory, we show that hallucination is inevitable for computable LLMs if the ground truth function is any computable function. Since the formal world is a part of the real world, we further conclude that it is impossible to eliminate hallucination in the real world LLMs. Using the formal world framework, we discuss the possible mechanisms and effectiveness of existing hallucination mitigators and discuss practical implications that our theoretical results have on the deployment of LLMs in the real world. We emphasize that since hallucination is inevitable, rigorous study of the safety of LLMs is critical.
причём:
...hallucination is inevitable for any computable LLM, regardless of model architecture, learning algorithms, prompting techniques, or training data.