Efficient slot labelling
This work addresses the problem of high computational costs for industry applications in dialogue systems, representing an incremental improvement in efficiency.
The paper tackles the computational inefficiency of pre-trained language models in slot labeling for dialogue systems by proposing a lightweight method that matches or exceeds state-of-the-art performance with nearly 10x fewer trainable parameters.
Slot labelling is an essential component of any dialogue system, aiming to find important arguments in every user turn. Common approaches involve large pre-trained language models (PLMs) like BERT or RoBERTa, but they face challenges such as high computational requirements and dependence on pre-training data. In this work, we propose a lightweight method which performs on par or better than the state-of-the-art PLM-based methods, while having almost 10x less trainable parameters. This makes it especially applicable for real-life industry scenarios.