Towards Linguistically Informed Multi-Objective Pre-Training for Natural Language Inference
This work addresses the need for more efficient models in natural language processing, though it is incremental as it builds on existing pre-training methods.
The paper tackles the problem of improving Natural Language Inference by introducing a linguistically enhanced pre-training method for transformers, achieving competitive results with a significant performance boost for smaller models.
We introduce a linguistically enhanced combination of pre-training methods for transformers. The pre-training objectives include POS-tagging, synset prediction based on semantic knowledge graphs, and parent prediction based on dependency parse trees. Our approach achieves competitive results on the Natural Language Inference task, compared to the state of the art. Specifically for smaller models, the method results in a significant performance boost, emphasizing the fact that intelligent pre-training can make up for fewer parameters and help building more efficient models. Combining POS-tagging and synset prediction yields the overall best results.