English

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

Artificial Intelligence 2024-09-09 v3 Computation and Language Machine Learning

Abstract

Can LLMs consistently improve their previous outputs for better results? For this to be true, LLMs would need to be better at discriminating among previously-generated alternatives, than generating initial responses. We explore the validity of this hypothesis in practice. We first formulate a unified framework that allows us to compare the generative and discriminative capability of any model on any task. In our resulting experimental analysis of several open-source and industrial LLMs, we observe that models are not reliably better at discriminating among previously-generated alternatives than generating initial responses. This finding challenges the notion that LLMs may be able to enhance their performance only through their own judgment.

Keywords

Cite

@article{arxiv.2404.04298,
  title  = {SELF-[IN]CORRECT: LLMs Struggle with Discriminating Self-Generated Responses},
  author = {Dongwei Jiang and Jingyu Zhang and Orion Weller and Nathaniel Weir and Benjamin Van Durme and Daniel Khashabi},
  journal= {arXiv preprint arXiv:2404.04298},
  year   = {2024}
}
R2 v1 2026-06-28T15:45:27.209Z