Multi-hop Selector Network for Multi-turn Response Selection in Retrieval-based Chatbots

Yuan, Chunyuan; Zhou, Wei; Li, Mingming; Lv, Shangwen; Zhu, Fuqing; Han, Jizhong; Hu, Songlin

doi:10.18653/v1/d19-1011

Cited by 111 publications

(126 citation statements)

References 21 publications

Supporting

Mentioning

120

Contrasting

Order By: Relevance

“…In the response selection, Zhang et al [46] propose a deep utterance aggregation model where the last utterance representation refines the preceding utterances to obtain the turns-aware representation and then the self-attention mechanism is applied among utterances. Yuan et al [42] propose a multi-hop selector network where the latter parts of the dialogue context are used as the query to select the relevant utterances on both the word-and utterance-level. As the knowledge sometimes contains a lot of redundant entries in knowledge-grounded conversation, Lian et at.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Learning to Detect Relevant Contexts and Knowledge for Response Selection in Retrieval-based Dialogue Systems

Hua

Zhao

Tao

et al. 2020

Proceedings of the 29th ACM International Conference on Information &Amp; Knowledge Management

View full text Add to dashboard Cite

Recently, knowledge-grounded conversations in the open domain gain great attention from researchers. Existing works on retrievalbased dialogue systems have paid tremendous efforts to utilize neural networks to build a matching model, where all of the context and knowledge contents are used to match the response candidate with various representation methods. Actually, different parts of the context and knowledge are differentially important for recognizing the proper response candidate, as many utterances are useless due to the topic shift. Those excessive useless information in the context and knowledge can influence the matching process and leads to inferior performance. To address this problem, we propose a multi-turn Response Selection Model that can Detect the relevant parts of the Context and Knowledge collection (RSM-DCK). Our model first uses the recent context as a query to pre-select relevant parts of the context and knowledge collection at the word-level and utterance-level semantics. Further, the response candidate interacts with the selected context and knowledge collection respectively. In the end, the fused representation of the context and response candidate is utilized to post-select the relevant parts of the knowledge collection for matching with more confidence. We test our proposed model on two benchmark datasets. Evaluation results indicate that our model achieves better performance than the existing methods, and can effectively detect the relevant context and knowledge for response selection.

show abstract

Section: Related Workmentioning

confidence: 99%

“…We set initial value of to 0.5 and make it update during the training. Following the MSN [42], we also perform multi-hop content selections to enhance the robustness of the selection module. Specifically, we first feed each of the latest utterances (namely…”

Section: Context Selectormentioning

confidence: 99%

Learning to Detect Relevant Contexts and Knowledge for Response Selection in Retrieval-based Dialogue Systems

Hua

Zhao

Tao

et al. 2020

Proceedings of the 29th ACM International Conference on Information &Amp; Knowledge Management

View full text Add to dashboard Cite

show abstract

“…We use the pyserini 17 implementation of QL and RM3, and use the context as query and candidate responses as candidate documents. In addition, we compare BERT against strong neural baselines for the task: DAM [62] 18 , and MSN [57] 19 , which are interaction-based methods that learn interactions between the utterances in the context and the response with attention and multi-hop selectors, respectively. We fine-tune the hyperparameters for the baseline models (QL, RM3, DAM, and MSN) using the validation set.…”

Section: Implementation Detailsmentioning

confidence: 99%

What does BERT know about books, movies and music? Probing BERT for Conversational Recommendation

Penha

Hauff

2020

Fourteenth ACM Conference on Recommender Systems

View full text Add to dashboard Cite

Heavily pre-trained transformer models such as BERT have recently shown to be remarkably powerful at language modelling by achieving impressive results on numerous downstream tasks. It has also been shown that they are able to implicitly store factual knowledge in their parameters after pre-training. Understanding what the pre-training procedure of LMs actually learns is a crucial step for using and improving them for Conversational Recommender Systems (CRS). We first study how much off-the-shelf pre-trained BERT "knows" about recommendation items such as books, movies and music. In order to analyze the knowledge stored in BERT's parameters, we use different probes (i.e., tasks to examine a trained model regarding certain properties) that require different types of knowledge to solve, namely content-based and collaborative-based. Content-based knowledge is knowledge that requires the model to match the titles of items with their content information, such as textual descriptions and genres. In contrast, collaborative-based knowledge requires the model to match items with similar ones, according to community interactions such as ratings. Both are important types of knowledge for a CRS-a system that should ideally be able to explain recommendations, match a user's textual description of their information need to relevant items, and elucidate a user's interests. We resort to BERT's Masked Language Modelling (MLM) head to probe its knowledge about the genre of items, with cloze style prompts. In addition, we employ BERT's Next Sentence Prediction (NSP) head and representations' similarity (SIM) to compare relevant and non-relevant search and recommendation query-document inputs to explore whether BERT can, without any fine-tuning, rank relevant items first. The insights we gain from these probes help us understand BERT's limitations and strengths for conversational recommendation. Finally, we study how BERT performs in a conversational recommendation downstream task. To this end, we fine-tune BERT to act as a retrieval-based conversational recommender system. Overall, our analyses and experiments show that: (i) BERT has knowledge stored in its parameters about the content of books, movies and music; (ii) it has more content-based knowledge than collaborative-based knowledge; and (iii) fails on conversational recommendation when faced with adversarial data.

show abstract

“…To mitigate this problem, two recently proposed notable studies are: Knowledge Enhanced Hybrid Neural Network (KEHNN) [32] and Multi-hop Selector Network (MSN) [31]. KEHNN leverages prior knowledge such as key phrases tagged in advance, to identify useful information in the dialogue context and performs matching with three interaction matrices.…”

Section: ) Knowledge Enhanced Hybrid Neural Networkmentioning

confidence: 99%

Exploiting Text Matching Techniques for Knowledge-Grounded Conversation

2020

View full text Add to dashboard Cite

Knowledge-grounded conversation models aim at generating informative responses for the given dialogue context, based on external knowledge. To generate an informative and context-coherent response, it is important to conjugate dialogue context and external knowledge in a balanced manner. However, existing studies have paid less attention to finding appropriate knowledge sentences from external knowledge sources than to generating proper sentences with correct dialogue acts. In this paper, we propose two knowledge selection strategies: 1) Reduce-Match and 2) Match-Reduce and explore several neural knowledge-grounded conversation models based on each strategy. Models based on Reduce-Match strategy first distill the whole dialogue context into a single vector with salient features preserved and then compare this context vector with the representation of knowledge sentences to predict a relevant knowledge sentence. Models based on Match-Reduce strategy first match every turn of the context with knowledge sentences to capture fine-grained interactions and aggregate them while minimizing information loss to predict the knowledge sentence. Experimental results show that conversation models using each of our knowledge selection strategies outperform the competitive baselines not only in terms of knowledge selection accuracy but also in response generation performance. Our best model based on Match-Reduce outperforms the baselines in the comparative studies with the Wizard of Wikipedia dataset. Also, our best model based on Reduce-Match outperforms them with the CMU Document Grounded Conversations dataset.

show abstract

Multi-hop Selector Network for Multi-turn Response Selection in Retrieval-based Chatbots

Cited by 111 publications

References 21 publications

Learning to Detect Relevant Contexts and Knowledge for Response Selection in Retrieval-based Dialogue Systems

Learning to Detect Relevant Contexts and Knowledge for Response Selection in Retrieval-based Dialogue Systems

What does BERT know about books, movies and music? Probing BERT for Conversational Recommendation

Exploiting Text Matching Techniques for Knowledge-Grounded Conversation

Contact Info

Product

Resources

About