CO3: Low-resource Contrastive Co-training for Generative Conversational Query Rewrite
This work addresses the challenge of expensive gold data and sensitivity to noise and style shifts in conversational AI, offering an incremental improvement for low-resource query rewrite tasks.
The paper tackles the problem of generative conversational query rewrite in low-resource settings by proposing a contrastive co-training method that uses unlabeled data to improve robustness to noise and language style shifts, achieving superior performance in few-shot and zero-shot scenarios with verified generalization ability.
Generative query rewrite generates reconstructed query rewrites using the conversation history while rely heavily on gold rewrite pairs that are expensive to obtain. Recently, few-shot learning is gaining increasing popularity for this task, whereas these methods are sensitive to the inherent noise due to limited data size. Besides, both attempts face performance degradation when there exists language style shift between training and testing cases. To this end, we study low-resource generative conversational query rewrite that is robust to both noise and language style shift. The core idea is to utilize massive unlabeled data to make further improvements via a contrastive co-training paradigm. Specifically, we co-train two dual models (namely Rewriter and Simplifier) such that each of them provides extra guidance through pseudo-labeling for enhancing the other in an iterative manner. We also leverage contrastive learning with data augmentation, which enables our model pay more attention on the truly valuable information than the noise. Extensive experiments demonstrate the superiority of our model under both few-shot and zero-shot scenarios. We also verify the better generalization ability of our model when encountering language style shift.