`Generalization is hallucination' through the lens of tensor completions

Abstract
In this short position paper, we introduce tensor completions and artifacts and make the case that they are a useful theoretical framework for understanding certain types of hallucinations and generalizations in language models.
View on arXiv@article{wong2025_2502.17305, title={ `Generalization is hallucination' through the lens of tensor completions }, author={ Liang Ze Wong }, journal={arXiv preprint arXiv:2502.17305}, year={ 2025 } }
Comments on this paper