Transformer encoder-decoder models have achieved great performance in dialogue generation tasks, however, their inability to process long dialogue history often leads to truncation of the context To address this problem, we propose a novel memory-augmented transformer that is compatible with existing pre-trained encoder-decoder models and enables efficient preservation of the dialogue history information. By incorporating a separate memory module alongside the pre-trained transformer, the model can effectively interchange information between the memory states and the current input context. We evaluate our model on three dialogue datasets and two language modeling datasets. Experimental results show that our method has achieved superior efficiency and performance compared to other pre-trained Transformer baselines.
Stateful Memory-Augmented Transformers for Efficient Dialogue Modeling
A memory-augmented Transformer model improves efficiency and performance in dialogue modeling by preserving history information without truncating sequences.
- Year
- 2022
- Venue
- arXiv 2022
- Authors
- 2
- Hosting
- Abstract onlyARXIV-DEFAULT
Cite
Notes
Only stored in your browser.
Attribution
- Abstract & full text
- arxiv.org/abs/2209.07634v2ARXIV-DEFAULT
- TL;DR
- Semantic Scholar