ParaQA: A Question Answering Dataset with Paraphrase Responses for Single-Turn Conversation

Kacupaj, Endri; Banerjee, Barshana; Singh, K. M.; Lehmann, Jens

doi:10.1007/978-3-030-77385-4_36

Cited by 13 publications

(14 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Following, multiple datasets were released to spur the community to apply end-to-end learning (Kacupaj et al, 2020(Kacupaj et al, , 2021aBiswas et al, 2021). Kacupaj et al (2021c) introduced VOGUE, an end-to-end model based on a dual encoder-decoder architecture.…”

Section: Modelmentioning

confidence: 99%

Proceedings of the Workshop on Structured and Unstructured Knowledge Integration (SUKI)

2022

View full text Add to dashboard Cite

As the demands for large-scale information processing have grown, knowledge graph-based approaches have gained prominence for representing general and domain knowledge. The development of such general representations is essential, particularly in domains such as manufacturing which intelligent processes and adaptive education can enhance. Despite the continuous accumulation of text in these domains, the lack of structured data has created information extraction and knowledge transfer barriers. In this paper, we report on work towards developing robust knowledge graphs based upon entity and relation data for both commercial and educational uses. To create the FabKG (Manufacturing knowledge graph), we have utilized textbook index words, research paper keywords, FabNER (manufacturing NER), to extract a sub knowledge base contained within Wikidata. Moreover, we propose a novel crowdsourcing method for KG creation by leveraging student notes, which contain invaluable information but are not captured as meaningful information, excluding their use in personal preparation for learning and written exams. We have created a knowledge graph containing 65000+ triples using all data sources. We have also shown the use case of domain-specific question answering and expression/formula-based question answering for educational purposes.

show abstract

Section: Modelmentioning

confidence: 99%

Proceedings of the Workshop on Structured and Unstructured Knowledge Integration (SUKI)

2022

View full text Add to dashboard Cite

show abstract

“…Evaluation Metrics We use the same metrics as employed by the authors of the three existing datasets (Kacupaj et al, , 2021aBiswas et al, 2021) on the previously mentioned baselines. The BLEU score, as defined by (Papineni et al, 2002), analyzes the co-occurrences of ngrams in the reference and the proposed responses.…”

Section: Model For Comparisonmentioning

confidence: 99%

“…ules (Unger et al, 2012). With the increasing popularity of intelligent personal assistants (e.g., Alexa, Siri), the research focus has been shifted to conversational question answering over KGs (ConvQA) that involve single-turn/multi-turn dialogues (Kacupaj et al, 2021a).…”

Section: Introductionmentioning

confidence: 99%

“…For the answer verbalization task, the system has to verbalize the answer to convey not only the information requested by the user but also additional characteristics that indicate how the answer was determined. In our exemplary question (from Kacupaj et al (2021a) dataset), a verbalized response would look like, "HBO owns 38 television shows." or "There are 38 TV shows whose owner is HBO.".…”

Section: Introductionmentioning

confidence: 99%

“…Both answers allow the user to verify that the system retrieved the total number of TV shows owned by HBO. In the literature, there exist empirical results showing that answer verbalization quantitatively and qualitatively improves the ability to understand the answer (Kacupaj et al, 2021a). However, it remains an open question -How can we verbalize an answer, given a logical form and an input question.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

VOGUE: Answer Verbalization through Multi-Task Learning

Kacupaj¹,

Premnadh²,

Singh³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

In recent years, there have been significant developments in Question Answering over Knowledge Graphs (KGQA). Despite all the notable advancements, current KGQA systems only focus on answer generation techniques and not on answer verbalization. However, in real-world scenarios (e.g., voice assistants such as Alexa, Siri, etc.), users prefer verbalized answers instead of a generated response. This paper addresses the task of answer verbalization for (complex) question answering over knowledge graphs. In this context, we propose a multi-task-based answer verbalization framework: VOGUE (Verbalization thrOuGh mUlti-task lEarning). The VOGUE framework attempts to generate a verbalized answer using a hybrid approach through a multitask learning paradigm. Our framework can generate results based on using questions and queries as inputs concurrently. VOGUE comprises four modules that are trained simultaneously through multi-task learning. We evaluate our framework on all existing datasets for answer verbalization, and it outperforms all current baselines on both BLEU and ME-TEOR scores as evaluation metric.

show abstract