Collapse of Self-trained Language Models

arXiv:2404.02305v1 Announce Type: new
Abstract: In various fields of knowledge creation, including science, new ideas often build on pre-existing information. In this work, we explore this concept within the context of language models. Specifically, we explore the potential of self-training models on their own outputs, akin to how humans learn and build on their previous thoughts and actions. While this approach is intuitively appealing, our research reveals its practical limitations. We find that extended self-training of the GPT-2 model leads …

abstract arxiv build concept context cs.ai cs.cl explore fields humans ideas information knowledge language language models learn science self-training training training models type work