LLMs, Reasoning and Plagiarism

Abstract

Recent reports claim that Large Language Models (LLMs) derive new science and exhibit human-level general intelligence. Such claims are entangled with two different narratives about what LLMs do: one in which they are an engine of synthesis that genuinely reasons to new knowledge, and one in which they retrieve and re-emit the work of others without attribution. In the scientific setting these are best understood as a contrast between \emph{reasoning} and \emph{plagiarism}. Finding where the truth lies between these two narratives is very challenging, as central components of the model -- the training data and the interaction transcript -- remain opaque. Thus claims of LLM reasoning do not satisfy Popper's refutability principle. We propose guidelines for transparency and reproducibility that will allow reasoning claims to be studied using the scientific method. The dominance of the reasoning narrative, we suggest, is in practice encouraging plagiarism in the scientific literature; we discuss what might be done about it.

arXiv commentThe authors explicitly reserve all rights in this work. No permission is granted for the reproduction, storage, or use of this document for the purpose of training artificial intelligence systems or for text and data mining (TDM), including but not limited to the generation of embeddings, summaries, or synthetic derivatives. Claude and Gemini were used in writing this manuscript