Proactive Human-Machine Conversation with Explicit Conversation Goal

Wu, Wenquan; Z, Guo; Zhou, Xiangyang; Wu, Hua; Zhang, Xiyuan; Lian, Rongzhong; Wang, Haifeng

doi:10.18653/v1/p19-1369

Cited by 133 publications

(131 citation statements)

References 19 publications

Supporting

Mentioning

130

Contrasting

Order By: Relevance

“…However, these models have no explicit high-level topics to guide multi-turn conversation generation, thus tending to generate less coherent dialogs. Recently, imposing goals on open-domain conversation generation models having attracted lots of research interests (Moon et al 2019;Li et al 2018;Tang et al 2019;Wu et al 2019) since it enables practical applications, e.g., recommendation of engaging entities. However, these models can just produce a dialog towards a single goal, instead of a goal sequence as done in this work.…”

Section: Related Workmentioning

confidence: 99%

“…The second one is how to generate an in-depth multi-turn conversation about a single topic for goal completion, which corresponds to intratopic coherence 2 . The capability of goal-sequence planning enables chatbots to conduct proactive open-domain conversations towards recommended topics (Moon et al 2019;Li et al 2018;Tang et al 2019;Wu et al 2019).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Knowledge Graph Grounded Goal Planning for Open-Domain Conversation Generation

Wang

Niu

et al. 2020

AAAI

Self Cite

View full text Add to dashboard Cite

Previous neural models on open-domain conversation generation have no effective mechanisms to manage chatting topics, and tend to produce less coherent dialogs. Inspired by the strategies in human-human dialogs, we divide the task of multi-turn open-domain conversation generation into two sub-tasks: explicit goal (chatting about a topic) sequence planning and goal completion by topic elaboration. To this end, we propose a three-layer Knowledge aware Hierarchical Reinforcement Learning based Model (KnowHRL). Specifically, for the first sub-task, the upper-layer policy learns to traverse a knowledge graph (KG) in order to plan a high-level goal sequence towards a good balance between dialog coherence and topic consistency with user interests. For the second sub-task, the middle-layer policy and the lower-layer one work together to produce an in-depth multi-turn conversation about a single topic with a goal-driven generation mechanism. The capability of goal-sequence planning enables chatbots to conduct proactive open-domain conversations towards recommended topics, which has many practical applications. Experiments demonstrate that our model outperforms state of the art baselines in terms of user-interest consistency, dialog coherence, and knowledge accuracy.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Knowledge Graph Grounded Goal Planning for Open-Domain Conversation Generation

Wang

Niu

et al. 2020

AAAI

Self Cite

View full text Add to dashboard Cite

show abstract

“…DuConv (Wu et al, 2019b): a proactive conversation dataset with 29858 dialogs and 270399 utterances. The model mainly plays the role of a leading player assigned with an explicit goal, a knowledge path comprised of two topics, and is provided with knowledge related to these two topics.…”

Section: Recurrent Knowledge Interactive Decodermentioning

confidence: 99%

“…While in dataset DuConv, a Chinese dialogue dataset with structured knowledge, we compare to the baselines referred in (Wu et al, 2019b) that consists of retrieval-based models as well as generation-based models.…”

Section: Comparison Approachesmentioning

confidence: 99%

Generating Informative Conversational Response using Recurrent Knowledge-Interaction and Knowledge-Copy

Lin¹,

Jian²,

He³

et al. 2020

Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics

View full text Add to dashboard Cite

Knowledge-driven conversation approaches have achieved remarkable research attention recently. However, generating an informative response with multiple relevant knowledge without losing fluency and coherence is still one of the main challenges. To address this issue, this paper proposes a method that uses recurrent knowledge interaction among response decoding steps to incorporate appropriate knowledge. Furthermore, we introduce a knowledge copy mechanism using a knowledge-aware pointer network to copy words from external knowledge according to knowledge attention distribution. Our joint neural conversation model which integrates recurrent Knowledge-Interaction and knowledge Copy (KIC) performs well on generating informative responses. Experiments demonstrate that our model with fewer parameters yields significant improvements over competitive baselines on two datasets Wizardof-Wikipedia(average Bleu +87%; abs.:0.034) and DuConv(average Bleu +20%; abs.:0.047) with different knowledge formats (textual & structured) and different languages (English & Chinese).

show abstract

“…We extend SRL to fit the conversational scenario by allowing SRL parsers to search for potential arguments over the whole conversation. As there is no publicly available data with paragraph-level SRL annotations, we directly annotate inter-and cross-utterance arguments for predicates on a public dialogue dataset, Duconv (Wu et al, 2019) 1 . Specifically, we annotated 3,000 dialogue sessions, including 33,673 predicates in 27,198 utterances.…”

Section: Conversational Srlmentioning

confidence: 99%

Semantic Role Labeling Guided Multi-turn Dialogue ReWriter

Tan

Song

et al. 2020

Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)

View full text Add to dashboard Cite

For multi-turn dialogue rewriting, the capacity of effectively modeling the linguistic knowledge in dialog context and getting rid of the noises is essential to improve its performance. Existing attentive models attend to all words without prior focus, which results in inaccurate concentration on some dispensable words. In this paper, we propose to use semantic role labeling (SRL), which highlights the core semantic information of who did what to whom, to provide additional guidance for the rewriter model. Experiments show that this information significantly improves a RoBERTa-based model that already outperforms previous stateof-the-art systems.

show abstract

Proactive Human-Machine Conversation with Explicit Conversation Goal

Abstract: arXiv:1906.05572v1 [cs.CL]

Cited by 133 publications

References 19 publications

Knowledge Graph Grounded Goal Planning for Open-Domain Conversation Generation

Knowledge Graph Grounded Goal Planning for Open-Domain Conversation Generation

Generating Informative Conversational Response using Recurrent Knowledge-Interaction and Knowledge-Copy

Semantic Role Labeling Guided Multi-turn Dialogue ReWriter

Contact Info

Product

Resources

About