Question Rewriting? Assessing Its Importance for Conversational Question Answering

Raposo, Gonçalo; Ribeiro, Rui; Martins, Bruno; Coheur, Luísa

doi:10.1007/978-3-030-99739-7_23

Cited by 11 publications

(11 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Question rewriting. We choose a very recent T5-based rewriting model [36]. The method is trained on the Canard question rewriting benchmark [9].…”

Section: Baselinesmentioning

confidence: 99%

“…One line of work that suggests a more general approach to the ConvQA problem aims to create a selfcontained question from the incomplete utterance that can be answered by standalone QA systems [9,36,51,52,56]. Such approaches can either take the form of question rewriting, which generates a complete question from scratch [36,51], or question resolution, which adds relevant terms from the conversational history to the question [52]. Question rewriting can entail unnecessary complexity since recent QA systems may not require the question to have perfect syntax or grammar [15,16].…”

Section: Related Workmentioning

confidence: 99%

“…For example, representations of the conversational context, like KB subgraphs or text passages, are often specifically modeled for the knowledge repository at hand, making these heterogeneous sources apparently incompatible. Methods for question rewriting [9,36,51] and question resolution [21,52] convert short user utterances into full-fledged questions where the intent is made completely explicit. However, this adds major complexity and may lose valuable cues from the conversation flow.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Conversational Question Answering on Heterogeneous Sources

Christmann¹,

Roy²,

Weikum³

2022

Preprint

View full text Add to dashboard Cite

Conversational question answering (ConvQA) tackles sequential information needs where contexts in follow-up questions are left implicit. Current ConvQA systems operate over homogeneous sources of information: either a knowledge base (KB), or a text corpus, or a collection of tables. This paper addresses the novel issue of jointly tapping into all of these together, this way boosting answer coverage and confidence. We present Convinse, an end-to-end pipeline for ConvQA over heterogeneous sources, operating in three stages: i) learning an explicit structured representation of an incoming question and its conversational context, ii) harnessing this framelike representation to uniformly capture relevant evidences from KB, text, and tables, and iii) running a fusion-in-decoder model to generate the answer. We construct and release the first benchmark, ConvMix, for ConvQA over heterogeneous sources, comprising 3000 real-user conversations with 16000 questions, along with entity annotations, completed question utterances, and question paraphrases. Experiments demonstrate the viability and advantages of our method, compared to state-of-the-art baselines. CCS CONCEPTS• Information systems → Question answering.

show abstract

“…Question rewriting. We choose a very recent T5-based rewriting model [36]. The method is trained on the Canard question rewriting benchmark [9].…”

Section: Baselinesmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Conversational Question Answering on Heterogeneous Sources

Christmann¹,

Roy²,

Weikum³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…To this end, several attempts have been made on developing end-to-end CQA models with dialogue history tracking (Qu et al, 2019;Qiu et al, 2021). Another group of works emphasizes the importance of query rewriting in CQA Raposo et al, 2022;Kim et al, 2021), which generates self-contained questions for performing single-turn QA. In addition, beyond simply focusing on one kind of information source, it has received increasing attentions to investigate CQA over hetergeneous sources (Li et al, 2022b;Christmann et al, 2022).…”

Section: Related Workmentioning

confidence: 99%

PACIFIC: Towards Proactive Conversational Question Answering over Tabular and Textual Data in Finance

Yang¹,

Lei²,

Zhang³

et al. 2022

Preprint

View full text Add to dashboard Cite

To facilitate conversational question answering (CQA) over hybrid contexts in finance, we present a new dataset, named PACIFIC. Compared with existing CQA datasets, PA-CIFIC exhibits three key features: (i) proactivity, (ii) numerical reasoning, and (iii) hybrid context of tables and text. A new task is defined accordingly to study Proactive Conversational Question Answering (PCQA), which combines clarification question generation and CQA. In addition, we propose a novel method, namely UniPCQA, to adapt a hybrid format of input and output content in PCQA into the Seq2Seq problem, including the reformulation of the numerical reasoning process as code generation. UniPCQA performs multitask learning over all sub-tasks in PCQA and incorporates a simple ensemble strategy to alleviate the error propagation issue in the multitask learning by cross-validating top-k sampled Seq2Seq outputs. We benchmark the PACIFIC dataset with extensive baselines and provide comprehensive evaluations on each sub-task of PCQA.

show abstract

“…However, it is also known to be quite challenging since it requires capturing all the information over the current question, previous conversation, and the given paragraph. To tackle this problem, a considerable amount of work focuses mainly on developing a model architecture for ConvQA (Qu et al, 2019a,b;Huang et al, 2019;Chen et al, 2020;Kim et al, 2021;Qiu et al, 2021;Raposo et al, 2022).…”

Section: Introductionmentioning

confidence: 99%

Realistic Conversational Question Answering with Answer Selection based on Calibrated Confidence and Uncertainty Measurement

Jeong,

Baek,

Hwang

et al. 2023

Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics

View full text Add to dashboard Cite

Conversational Question Answering (ConvQA) models aim at answering a question with its relevant paragraph and previous question-answer pairs that occurred during conversation multiple times. To apply such models to a real-world scenario, some existing work uses predicted answers, instead of unavailable ground-truth answers, as the conversation history for inference. However, since these models usually predict wrong answers, using all the predictions without filtering significantly hampers the model performance. To address this problem, we propose to filter out inaccurate answers in the conversation history based on their estimated confidences and uncertainties from the Con-vQA model, without making any architectural changes. Moreover, to make the confidence and uncertainty values more reliable, we propose to further calibrate them, thereby smoothing the model predictions. We validate our models, Answer Selection-based realistic Conversation Question Answering, on two standard ConvQA datasets, and the results show that our models significantly outperform relevant baselines.

show abstract

Question Rewriting? Assessing Its Importance for Conversational Question Answering

Cited by 11 publications

References 9 publications

Conversational Question Answering on Heterogeneous Sources

Conversational Question Answering on Heterogeneous Sources

PACIFIC: Towards Proactive Conversational Question Answering over Tabular and Textual Data in Finance

Realistic Conversational Question Answering with Answer Selection based on Calibrated Confidence and Uncertainty Measurement

Contact Info

Product

Resources

About