Counterfactual Language Model Adaptation for Suggesting Phrases
This addresses the issue of inefficient text entry on mobile devices by improving suggestion acceptance, though it is incremental as it builds on existing language models.
The paper tackled the problem of generating phrase suggestions that writers are more likely to accept, rather than just predicting word frequency, and found that a simple language model can improve acceptability in a counterfactual offline setting.
Mobile devices use language models to suggest words and phrases for use in text entry. Traditional language models are based on contextual word frequency in a static corpus of text. However, certain types of phrases, when offered to writers as suggestions, may be systematically chosen more often than their frequency would predict. In this paper, we propose the task of generating suggestions that writers accept, a related but distinct task to making accurate predictions. Although this task is fundamentally interactive, we propose a counterfactual setting that permits offline training and evaluation. We find that even a simple language model can capture text characteristics that improve acceptability.