PIP: Parse-Instructed Prefix for Syntactically Controlled Paraphrase Generation
This work addresses the high computational cost for researchers and practitioners in NLP by providing an efficient alternative for syntactically controlled paraphrase generation, though it is incremental as it builds on prefix-tuning.
The paper tackles the problem of costly fine-tuning for syntactically controlled paraphrase generation by proposing PIP, a parameter-efficient method that reduces learnable parameters by 10 times while achieving significantly higher performance compared to existing prefix-tuning methods.
Syntactically controlled paraphrase generation requires language models to generate paraphrases for sentences according to specific syntactic structures. Existing fine-tuning methods for this task are costly as all the parameters of the model need to be updated during the training process. Inspired by recent studies on parameter-efficient learning, we propose Parse-Instructed Prefix (PIP), a novel adaptation of prefix-tuning to tune large pre-trained language models on syntactically controlled paraphrase generation task in a low-data setting with significantly less training cost. We introduce two methods to instruct a model's encoder prefix to capture syntax-related knowledge: direct initiation (PIP-Direct) and indirect optimization (PIP-Indirect). In contrast to traditional fine-tuning methods for this task, PIP is a compute-efficient alternative with 10 times less learnable parameters. Compared to existing prefix-tuning methods, PIP excels at capturing syntax control information, achieving significantly higher performance at the same level of learnable parameter count.