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

Lin, Xiexiong; Jian, Weiyu; He, Jianshan; Wang, Taifeng; Chu, Wei

doi:10.18653/v1/2020.acl-main.6

Cited by 50 publications

(24 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus, recently, external information in the form of structured knowledge graphs (KG) is introduced to enhance item representations by using rich entity information in KG (Chen et al, 2019;. While KGbased methods improve CR to some extent, they are still limited in (i) worse versatility resulted from a high cost of KG construction; and (ii) inadequate integration of knowledge and response generation (Lin et al, 2020). Given that, nowadays, users are greatly encouraged to share their consumption experience (e.g., restaurant, traveling, movie, etc.…”

Section: … Conversational Recommendationmentioning

confidence: 99%

RevCore: Review-Augmented Conversational Recommendation

Yu¹,

Bao²,

Yan³

et al. 2021

Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021

View full text Add to dashboard Cite

Existing conversational recommendation (CR) systems usually suffer from insufficient item information when conducted on short dialogue history and unfamiliar items. Incorporating external information (e.g., reviews) is a potential solution to alleviate this problem. Given that reviews often provide a rich and detailed user experience on different interests, they are potential ideal resources for providing high-quality recommendations within an informative conversation. In this paper, we design a novel end-to-end framework, namely, Review-augmented Conversational Recommender (RevCore), where reviews are seamlessly incorporated to enrich item information and assist in generating both coherent and informative responses. In detail, we extract sentiment-consistent reviews, perform review-enriched and entity-based recommendations for item suggestions, as well as use a review-attentive encoder-decoder for response generation. Experimental results demonstrate the superiority of our approach in yielding better performance on both recommendation and conversation responding. 1

show abstract

Section: … Conversational Recommendationmentioning

confidence: 99%

RevCore: Review-Augmented Conversational Recommendation

Yu¹,

Bao²,

Yan³

et al. 2021

Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021

View full text Add to dashboard Cite

show abstract

“…We leverage piece-based unstructured knowledge and focus on improving KS. Several previous publications focus on implicit KS to calculate a weight for each piece of knowledge and get the weighted sum of their representations [9,25,63]. Other studies focus on explicit KS [1,62], i.e., calculating a weight on each piece of knowledge and then directly sample a piece of knowledge with the highest weight.…”

Section: Knowledge-grounded Conversationmentioning

confidence: 99%

Initiative-Aware Self-Supervised Learning for Knowledge-Grounded Conversations

Chen

Ren

Chen

et al. 2021

Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval

View full text Add to dashboard Cite

In the knowledge-grounded conversation (KGC) task systems aim to produce more informative responses by leveraging external knowledge. KGC includes a vital part, knowledge selection, where conversational agents select the appropriate knowledge to be incorporated in the next response. Mixed initiative is an intrinsic feature of conversations where the user and the system can both take the initiative in suggesting new conversational directions. Knowledge selection can be driven by the user's initiative or by the system's initiative. For the former, the system usually selects knowledge according to the current user utterance that contains new topics or questions posed by the user; for the latter, the system usually selects knowledge according to the previously selected knowledge. No previous study has considered the mixed-initiative characteristics of knowledge selection to improve its performance.In this paper, we propose a mixed-initiative knowledge selection method (MIKe) for KGC, which explicitly distinguishes between user-initiative and system-initiative knowledge selection. Specifically, we introduce two knowledge selectors to model both of them separately, and design a novel initiative discriminator to discriminate the initiative type of knowledge selection at each conversational turn. A challenge for training MIKe is that we usually have no labels for indicating initiative. To tackle this challenge, we devise an initiative-aware self-supervised learning scheme that helps MIKe to learn to discriminate the initiative type via a self-supervised task. Experimental results on two datasets show that MIKe significantly outperforms state-of-the-art methods in terms of both automatic and human evaluations, indicating that it can select more appropriate knowledge and generate more informative and engaging responses.

show abstract

“…Li and Srikumar (2019) use first-order logic to constrain the output of neural layers. Other methods use symbolic knowledge in pretraining (Rosset et al, 2020;Tian et al, 2020) or use symbolic rules to constrain text generation (Lin et al, 2020;Li and Rush, 2020). Our method is different from these methods in that we directly neuralize the RE symbolic system, instead of having separate symbolic and neural systems.…”

Section: Related Workmentioning

confidence: 99%

Neuralizing Regular Expressions for Slot Filling

Jiang¹,

Zhang²,

Tu³

2021

Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing

View full text Add to dashboard Cite

Neural models and symbolic rules such as regular expressions have their respective merits and weaknesses. In this paper, we study the integration of the two approaches for the slot filling task by converting regular expressions into neural networks. Specifically, we first convert regular expressions into a special form of finite-state transducers, then unfold its approximate inference algorithm as a bidirectional recurrent neural model that performs slot filling via sequence labeling. Experimental results show that our model has superior zero-shot and few-shot performance and stays competitive when there are sufficient training data.

show abstract

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

Cited by 50 publications

References 18 publications

RevCore: Review-Augmented Conversational Recommendation

RevCore: Review-Augmented Conversational Recommendation

Initiative-Aware Self-Supervised Learning for Knowledge-Grounded Conversations

Neuralizing Regular Expressions for Slot Filling

Contact Info

Product

Resources

About