Common ground plays a crucial role in human communication and the grounding process helps to establish shared knowledge. However, common ground is also a heavily loaded term that may be interpreted in different ways depending on the context. The scope of common ground ranges from domain-specific and personal shared experiences to common sense knowledge. Representationally, common ground can be uni- or multi-modal, and static or dynamic. In this survey, we attempt to systematize different facets of common ground in dialogue and position it within the current landscape of NLP research that often relies on the usage of language models (LMs) and task-specific short-term interactions. We outline different dimensions of common ground and describe modeling approaches for several grounding tasks, discuss issues caused by the lack of common ground in human-LM interactions, and suggest future research directions. This survey serves as a roadmap of what to pay attention to when equipping a dialogue system with grounding capabilities and provides a summary of current research on grounding in dialogue, categorizing 448 papers and compiling a list of the available datasets.

Misunderstandings occur all the time in human conversation but deciding on when to ask for clarification is a challenging task for conversational systems that requires a balance between asking too many unnecessary questions and running the risk of providing incorrect information. This work investigates clarification identification based on the task and data from (Xu et al., 2019), reproducing their Transformer baseline and extending it by comparing pre-trained language model fine-tuning, prompt tuning and manual prompt engineering on the task of clarification identification. Our experiments show strong performance with LM and a prompt tuning approach with BERT and RoBERTa, outperforming standard LM fine-tuning, while manual prompt engineering with GPT-3.5 proved to be less effective, although informative prompt instructions have the potential of steering the model towards generating more accurate explanations for why clarification is needed.