STaR-GATE: Teaching Language Models to Ask Clarifying Questions

arXiv:2403.19154v1 Announce Type: new
Abstract: When prompting language models to complete a task, users often leave important aspects unsaid. While asking questions could resolve this ambiguity \citep[GATE;][]{li2023eliciting}, models often struggle to ask good questions. We explore a language model's ability to self-improve \citep[STaR;][]{zelikman2022star} by rewarding the model for generating useful questions -- a simple method we dub STaR-GATE. We generate a synthetic dataset of 25,500 unique persona-task prompts to simulate conversations between a pretrained language model -- the \texttt{Questioner} -- …

abstract arxiv cs.ai cs.cl explore gate good language language model language models prompting questions star struggle teaching type