We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.CL

Change to browse by:

cs

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Computation and Language

Title: Conversation Graph: Data Augmentation, Training and Evaluation for Non-Deterministic Dialogue Management

Abstract: Task-oriented dialogue systems typically rely on large amounts of high-quality training data or require complex handcrafted rules. However, existing datasets are often limited in size considering the complexity of the dialogues. Additionally, conventional training signal inference is not suitable for non-deterministic agent behaviour, i.e. considering multiple actions as valid in identical dialogue states. We propose the Conversation Graph (ConvGraph), a graph-based representation of dialogues that can be exploited for data augmentation, multi-reference training and evaluation of non-deterministic agents. ConvGraph generates novel dialogue paths to augment data volume and diversity. Intrinsic and extrinsic evaluation across three datasets shows that data augmentation and/or multi-reference training with ConvGraph can improve dialogue success rates by up to 6.4%.
Comments: Accepted at Transactions of Association of Computational Linguistics (to be presented at ACL 2021)
Subjects: Computation and Language (cs.CL)
Cite as: arXiv:2010.15411 [cs.CL]
  (or arXiv:2010.15411v2 [cs.CL] for this version)

Submission history

From: Milan Gritta [view email]
[v1] Thu, 29 Oct 2020 08:23:24 GMT (165kb,D)
[v2] Wed, 4 Nov 2020 14:22:50 GMT (167kb,D)

Link back to: arXiv, form interface, contact.