Corpus ID: 229371419

I like fish, especially dolphins: Addressing Contradictions in Dialogue Modeling

@article{Nie2020ILF,
  title={I like fish, especially dolphins: Addressing Contradictions in Dialogue Modeling},
  author={Yixin Nie and Mary Williamson and M. Bansal and Douwe Kiela and J. Weston},
  journal={ArXiv},
  year={2020},
  volume={abs/2012.13391}
}
To quantify how well natural language understanding models can capture consistency in a general conversation, we introduce the DialoguE COntradiction DEtection task (DECODE) and a new conversational dataset containing both human-human and human-bot contradictory dialogues. We then compare a structured utterance-based approach of using pre-trained Transformer models for contradiction detection with the typical unstructured approach. Results reveal that: (i) our newly collected dataset is notably… Expand

Figures and Tables from this paper

References

SHOWING 1-10 OF 56 REFERENCES
Evaluating Coherence in Dialogue Systems using Entailment
  • 23
  • PDF
Don't Say That! Making Inconsistent Dialogue Unlikely with Unlikelihood Training
  • 28
  • PDF
DialoGPT: Large-Scale Generative Pre-training for Conversational Response Generation
  • 158
  • PDF
ACUTE-EVAL: Improved Dialogue Evaluation with Optimized Questions and Multi-turn Comparisons
  • 38
  • PDF
Consistent Dialogue Generation with Self-supervised Feature Learning
  • 16
  • PDF
Towards Empathetic Open-domain Conversation Models: A New Benchmark and Dataset
  • 105
  • PDF
The Ubuntu Dialogue Corpus: A Large Dataset for Research in Unstructured Multi-Turn Dialogue Systems
  • 574
  • PDF
Augmenting Neural Response Generation with Context-Aware Topical Attention
  • 27
  • PDF
...
1
2
3
4
5
...