mlsanityData to 7 Oct 2026

How to Build Robust FAQ Chatbot with Controllable Question Generator?

arXiv 2112.03007 ↗First posted 18 Nov 2021cs.CL cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Many unanswerable adversarial questions fool the question-answer (QA) system with some plausible answers. Building a robust, frequently asked questions (FAQ) chatbot needs a large amount of diverse adversarial examples. Recent question generation methods are ineffective at generating many high-quality and diverse adversarial question-answer pairs from unstructured text. We propose the diversity controllable semantically valid adversarial attacker (DCSA), a high-quality, diverse, controllable method to generate standard and adversarial samples with a semantic graph. The fluent and semantically generated QA pairs fool our passage retrieval model successfully. After that, we conduct a study on the robustness and generalization of the QA model with generated QA pairs among different domains. We find that the generated data set improves the generalizability of the QA model to the new target domain and the robustness of the QA model to detect unanswerable adversarial questions.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 28 of its 36 references that are in the map. First posted on arXiv 18 Nov 2021.

Builds on

28 of 36 references.

References · 28

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  2. 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,266
  3. 2016SQuAD: 100,000+ Questions for Machine Comprehension of Textcited by 4,115
  4. 2019ALBERT: A Lite BERT for Self-supervised Learning of Language Representationscited by 2,772
  5. 2018Know What You Don't Know: Unanswerable Questions for SQuADcited by 1,538
  6. 2017Reading Wikipedia to Answer Open-Domain Questionscited by 777
  7. 2017Adversarial Examples for Evaluating Reading Comprehension Systemscited by 568
  8. 2020BERT-ATTACK: Adversarial Attack Against BERT Using BERTcited by 488
  9. 2018QuAC : Question Answering in Contextcited by 379
  10. 2019Multilingual Universal Sentence Encoder for Semantic Retrievalcited by 192
  11. 2016A Hierarchical Latent Variable Encoder-Decoder Model for Generating Dialoguescited by 174
  12. 2018Generating Informative and Diverse Conversational Responses via Adversarial Information Maximizationcited by 114
16 more
  1. 2017Deep Text Classification Can be Fooledcited by 105
  2. 2017Neural Question Generation from Text: A Preliminary Studycited by 78
  3. 2019MultiQA: An Empirical Investigation of Generalization and Transfer in Reading Comprehensioncited by 60
  4. 2016Generating Natural Questions About an Imagecited by 56
  5. 2018Harvesting Paragraph-Level Question-Answer Pairs from Wikipediacited by 49
  6. 2020A Controllable Model of Grounded Response Generationcited by 49
  7. 2018Learning Cross-Lingual Sentence Representations via a Multi-task Dual-Encoder Modelcited by 46
  8. 2020What do Models Learn from Question Answering Datasets?cited by 45
  9. 2020Asking Questions the Human Way: Scalable Question-Answer Generation from Text Corpuscited by 40
  10. 2017Question Answering and Question Generation as Dual Taskscited by 37
  11. 2018Improving Neural Question Generation using Answer Separationcited by 30
  12. 2018A Qualitative Comparison of CoQA, SQuAD 2.0 and QuACcited by 21
  13. 2019Learning to Answer by Learning to Ask: Getting the Best of GPT-2 and BERT Worldscited by 19
  14. 2018Zero-Shot Question Generation from Knowledge Graphs for Unseen Predicates and Entity Typescited by 17
  15. 2020Improving the Robustness of QA Models to Challenge Sets with Variational Question-Answer Pair Generationcited by 10
  16. 2020QnAMaker: Data to Bot in 2 Minutescited by 1

Built on by

Not cited yet.