REVIEW 5 cited by
Learning Discourse-level Diversity for Neural Dialog Models using Conditional Variational Autoencoders
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Signed reviews
read the original abstract
While recent neural encoder-decoder models have shown great promise in modeling open-domain conversations, they often generate dull and generic responses. Unlike past work that has focused on diversifying the output of the decoder at word-level to alleviate this problem, we present a novel framework based on conditional variational autoencoders that captures the discourse-level diversity in the encoder. Our model uses latent variables to learn a distribution over potential conversational intents and generates diverse responses using only greedy decoders. We have further developed a novel variant that is integrated with linguistic prior knowledge for better performance. Finally, the training procedure is improved by introducing a bag-of-word loss. Our proposed models have been validated to generate significantly more diverse responses than baseline approaches and exhibit competence in discourse-level decision-making.
Forward citations
Cited by 5 Pith papers
-
Variational Inference for Bird's Eye View Segmentation in Autonomous Driving
TVB combines a conditional variational autoencoder, normalizing flows, and attention-based fusion to produce bird's-eye-view segmentation from multiple car cameras, reporting small but consistent IoU gains on nuScenes...
-
TransSent: Towards Generation of Structured Sentences with Discourse Marker
TransSent generates a tail discourse from a head discourse and a discourse marker by treating the marker as a translation in embedding space, with new datasets and improved scores over baselines.
-
Mixture Content Selection for Diverse Sequence Generation
A mixture-of-experts content selector that masks different input tokens for each generated sequence improves diversity and accuracy in question generation and summarization.
-
Implicit Deep Latent Variable Models for Text Generation
Implicit (sample-based) variational posteriors with aggregated-posterior matching beat Gaussian VAEs on text generation and reduce posterior collapse.
-
Deep Learning Based Chatbot Models
A 2017 student report surveys over 70 chatbot papers and reports preliminary Transformer experiments suggesting the model underperforms seq2seq on dialogue while speaker-addressee conditioning changes response quality.
Discussion (0). Continue with ORCID to comment.