Dual Supervised Learning for Natural Language Understanding and Generation
This addresses a gap in NLP research by exploiting the duality between NLU and NLG, but it appears incremental as it builds on existing dual supervised learning concepts.
The paper tackled the problem of leveraging the dual relationship between natural language understanding (NLU) and generation (NLG) by proposing a new learning framework based on dual supervised learning, and preliminary experiments showed it boosts performance for both tasks.
Natural language understanding (NLU) and natural language generation (NLG) are both critical research topics in the NLP field. Natural language understanding is to extract the core semantic meaning from the given utterances, while natural language generation is opposite, of which the goal is to construct corresponding sentences based on the given semantics. However, such dual relationship has not been investigated in the literature. This paper proposes a new learning framework for language understanding and generation on top of dual supervised learning, providing a way to exploit the duality. The preliminary experiments show that the proposed approach boosts the performance for both tasks.