Look before you Hop: Conversational Question Answering over Knowledge Graphs Using Judicious Context Expansion

Christmann, Philipp; Roy, Rishiraj Saha; Abujabal, Abdalghani; Singh, Jyotsna; Weikum, Gerhard

doi:10.1145/3357384.3358016

Cited by 78 publications

(115 citation statements)

References 32 publications

(55 reference statements)

Supporting

Mentioning

114

Contrasting

Order By: Relevance

“…The evaluation of interactive systems requires real user sessions, as explained in §3. Using a prototype INTSUMM system, described in §5.1, we conducted several cycles of session collection which uncovered multiple user-related challenges, in line with previous work on user task design (Christmann et al, 2019;Roit et al, 2020;Zuccon et al, 2013). In particular, recruited users may make undue interactions due to insincere or experimental behavior, yielding noisy sessions that do not reflect realistic system use.…”

Section: Session Collectionmentioning

confidence: 98%

Extending Multi-Document Summarization Evaluation to the Interactive Setting

Shapira¹,

Pasunuru²,

Ronen³

et al. 2021

Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Langua

View full text Add to dashboard Cite

Allowing users to interact with multidocument summarizers is a promising direction towards improving and customizing summary results. Different ideas for interactive summarization have been proposed in previous work but these solutions are highly divergent and incomparable. In this paper, we develop an end-to-end evaluation framework for interactive summarization, focusing on expansion-based interaction, which considers the accumulating information along a user session. Our framework includes a procedure of collecting real user sessions, as well as evaluation measures relying on summarization standards, but adapted to reflect interaction. All of our solutions and resources are available publicly as a benchmark, allowing comparison of future developments in interactive summarization, and spurring progress in its methodological evaluation. We demonstrate the use of our framework by evaluating and comparing baseline implementations that we developed for this purpose, which will serve as part of our benchmark. Our extensive experimentation and analysis motivate the proposed evaluation framework design and support its viability.

show abstract

Section: Session Collectionmentioning

confidence: 98%

Extending Multi-Document Summarization Evaluation to the Interactive Setting

Shapira¹,

Pasunuru²,

Ronen³

et al. 2021

Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Langua

View full text Add to dashboard Cite

show abstract

“…3.1.1 Dataset. We constructed datasets for CQR based on conversational challenges in CSQA [9] and ConvQuestions [3]. CSQA is the largest dataset (200k dialogues, 1.6M turns) for conversational QA over KB (Wikidata with 12.8M entities).…”

Section: Experimentation 31 Setupmentioning

confidence: 99%

“…Considering originality of action-based CQR for general CQU, we present reconstructed datasets based on conversational challenges in Complex Sequential Question Answering (CSQA) [9] and ConvQuestions [3], typical multi-domain conversational QA, with actions summarized from query types and transformation patterns.…”

Section: Introductionmentioning

confidence: 99%

Learning to Generate Reformulation Actions for Scalable Conversational Query Understanding

Zhu²,

Geng³

et al. 2020

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

View full text Add to dashboard Cite

The ability of conversational query understanding (CQU) is indispensable to multi-turn QA. However, existing methods are datadriven and expensive to extend to new conversation domains, or under specific frameworks and hard to apply to other underlying QA technologies. We propose a novel contextual query reformulation (CQR) module based on reformulation actions for general CQU. The actions are domain-independent and scalable, since they capture syntactic regularities of conversations. For action generation, we propose a multi-task learning framework enhanced by coreference resolution, and introduce grammar constraints into the decoding process. Then CQR synthesizes standalone queries based on the actions, which naturally adapts to original downstream technologies. Experiments on different CQU datasets suggest that action-based methods substantially outperform direct reformulation, and the proposed model performs the best among the methods.

show abstract

“…Recently, conversational Knowledge Base Question Answering (KBQA) has started to attract people's attention (Saha et al, 2018;Christmann et al, 2019;Guo et al, 2018;Shen et al, 2019). Motivated by real-world conversational applications, particularly personal assistants such as Apple Siri and Amazon Alexa, the task aims to answer questions over KBs in a conversational manner.…”

Section: Introductionmentioning

confidence: 99%

Modeling Transitions of Focal Entities for Conversational Knowledge Base Question Answering

Lan¹,

Jiang²

2021

Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Confer

View full text Add to dashboard Cite

Conversational KBQA is about answering a sequence of questions related to a KB. Follow-up questions in conversational KBQA often have missing information referring to entities from the conversation history. In this paper, we propose to model these implied entities, which we refer to as the focal entities of the conversation. We propose a novel graph-based model to capture the transitions of focal entities and apply a graph neural network to derive a probability distribution of focal entities for each question, which is then combined with a standard KBQA module to perform answer ranking. Our experiments on two datasets demonstrate the effectiveness of our proposed method.

show abstract

Look before you Hop: Conversational Question Answering over Knowledge Graphs Using Judicious Context Expansion

Cited by 78 publications

References 32 publications

Extending Multi-Document Summarization Evaluation to the Interactive Setting

Extending Multi-Document Summarization Evaluation to the Interactive Setting

Learning to Generate Reformulation Actions for Scalable Conversational Query Understanding

Modeling Transitions of Focal Entities for Conversational Knowledge Base Question Answering

Contact Info

Product

Resources

About