SELF-[IN]CORRECT: LLMs Struggle with Refining Self-Generated Responses

arXiv:2404.04298v1 Announce Type: cross
Abstract: Can LLMs continually improve their previous outputs for better results? An affirmative answer would require LLMs to be better at discriminating among previously-generated alternatives, than generating initial responses. We explore the validity of this hypothesis in practice. We first introduce a unified framework that allows us to compare the generative and discriminative capability of any model on any task. Then, in our resulting experimental analysis of several LLMs, we do not observe the performance of …

abstract arxiv cs.ai cs.cl cs.lg explore framework generated hypothesis llms practice responses results struggle type