Generative Zero-Shot Prompt Learning for Cross-Domain Slot Filling with Inverse Prompting
This work addresses the problem of poor generalization and robustness in slot filling for natural language processing, offering a domain-specific incremental advance.
The paper tackles zero-shot cross-domain slot filling by proposing a generative prompt learning framework with inverse prompting and efficient prompt-tuning, achieving a 13.44% F1 improvement on unseen slots.
Zero-shot cross-domain slot filling aims to transfer knowledge from the labeled source domain to the unlabeled target domain. Existing models either encode slot descriptions and examples or design handcrafted question templates using heuristic rules, suffering from poor generalization capability or robustness. In this paper, we propose a generative zero-shot prompt learning framework for cross-domain slot filling, both improving generalization and robustness than previous work. Besides, we introduce a novel inverse prompting strategy to distinguish different slot types to avoid the multiple prediction problem, and an efficient prompt-tuning strategy to boost higher performance by only training fewer prompt parameters. Experiments and analysis demonstrate the effectiveness of our proposed framework, especially huge improvements (+13.44% F1) on the unseen slots.