Self-Contrast: Better Reflection Through Inconsistent Solving Perspectives

arXiv:2401.02009v2 Announce Type: replace
Abstract: The reflection capacity of Large Language Model (LLM) has garnered extensive attention. A post-hoc prompting strategy, e.g., reflexion and self-refine, refines LLM's response based on self-evaluated or external feedback. However, recent research indicates without external feedback, LLM's intrinsic reflection is unstable. Our investigation unveils that the key bottleneck is the quality of the self-evaluated feedback. We find LLMs often exhibit overconfidence or high randomness when self-evaluate, offering stubborn or inconsistent feedback, which causes poor reflection. …

abstract arxiv attention capacity contrast cs.ai cs.cl feedback however intrinsic investigation key language language model large language large language model llm perspectives prompting refine research strategy the key through type