Abstract
Task-oriented dialog models typically leverage complex neural architectures and large-scale, pre-trained Transformers to achieve state-of-the-art performance on popular natural language understanding benchmarks. However, these models frequently have in excess of tens of millions of parameters, making them impossible to deploy on-device where resource-efficiency is a major concern. In this work, we show that a simple convolutional model compressed with structured pruning achieves largely comparable results to BERT on ATIS and Snips, with under 100K parameters. Moreover, we perform acceleration experiments on CPUs, where we observe our multi-task model predicts intents and slots nearly 63x faster than even DistilBERT.- Anthology ID:
- 2020.nlp4convai-1.6
- Volume:
- Proceedings of the 2nd Workshop on Natural Language Processing for Conversational AI
- Month:
- July
- Year:
- 2020
- Address:
- Online
- Editors:
- Tsung-Hsien Wen, Asli Celikyilmaz, Zhou Yu, Alexandros Papangelis, Mihail Eric, Anuj Kumar, Iñigo Casanueva, Rushin Shah
- Venue:
- NLP4ConvAI
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 46–53
- Language:
- URL:
- https://aclanthology.org/2020.nlp4convai-1.6
- DOI:
- 10.18653/v1/2020.nlp4convai-1.6
- Cite (ACL):
- Ojas Ahuja and Shrey Desai. 2020. Accelerating Natural Language Understanding in Task-Oriented Dialog. In Proceedings of the 2nd Workshop on Natural Language Processing for Conversational AI, pages 46–53, Online. Association for Computational Linguistics.
- Cite (Informal):
- Accelerating Natural Language Understanding in Task-Oriented Dialog (Ahuja & Desai, NLP4ConvAI 2020)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-4/2020.nlp4convai-1.6.pdf
- Code
- oja/pruned-nlu
- Data
- ATIS, SNIPS