Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
119 tokens/sec
GPT-4o
56 tokens/sec
Gemini 2.5 Pro Pro
43 tokens/sec
o3 Pro
6 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Conversational Question Answering on Heterogeneous Sources (2204.11677v2)

Published 25 Apr 2022 in cs.IR and cs.CL

Abstract: 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 frame-like 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.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (59)
  1. DBpedia: A nucleus for a Web of open data. In Sem. Web.
  2. Reading Wikipedia to Answer Open-Domain Questions. In ACL.
  3. Open question answering over tables and text. In ICLR.
  4. HybridQA: A Dataset of Multi-Hop Question Answering over Tabular and Textual Data. In EMNLP.
  5. GraphFlow: Exploiting Conversation Flow with Graph Neural Networks for Conversational Machine Comprehension. In IJCAI.
  6. QuAC: Question answering in context. In EMNLP.
  7. Beyond NED: Fast and Effective Search Space Reduction for Complex Question Answering over Knowledge Bases. In WSDM.
  8. Look before you hop: Conversational question answering over knowledge graphs using judicious context expansion. In CIKM.
  9. Can You Unpack That? Learning to Rewrite Questions-in-Context. In EMNLP-IJCNLP.
  10. Dialog-to-action: conversational question answering over a large-scale knowledge base. In NeurIPS.
  11. Manish Gupta and Michael Bendersky. 2015. Information retrieval with verbose queries. In SIGIR.
  12. ManyModalQA: Modality Disambiguation and QA over Diverse Inputs. In AAAI, Vol. 34.
  13. Qsearch: Answering quantity queries from text. In ISWC.
  14. FlowQA: Grasping Flow in History for Conversational Machine Comprehension. In ICLR.
  15. Search-based neural structured learning for sequential question answering. In ACL.
  16. Gautier Izacard and Édouard Grave. 2021. Leveraging Passage Retrieval with Generative Models for Open Domain Question Answering. In EACL.
  17. Complex Temporal Question Answering on Knowledge Graphs. In CIKM.
  18. Conversational Entity Linking: Problem Definition and Datasets. In SIGIR.
  19. Conversational Question Answering over Knowledge Graphs with Transformer and Graph Attention Networks. In EACL.
  20. Conversational Question Answering over Passages by Leveraging Word Proximity Networks. In SIGIR.
  21. Reinforcement Learning from Reformulations in Conversational Question Answering over Knowledge Graphs. In SIGIR.
  22. Vineet Kumar and Sachindra Joshi. 2017. Incomplete follow-up question resolution using retrieval based sequence to sequence learning. In SIGIR.
  23. Yunshi Lan and Jing Jiang. 2021. Modeling Transitions of Focal Entities for Conversational Knowledge Base Question Answering. In ACL.
  24. Vladimir I Levenshtein et al. 1966. Binary codes capable of correcting deletions, insertions, and reversals. In Soviet physics doklady.
  25. BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension. In ACL.
  26. Open Domain Question Answering with A Unified Knowledge Interface. In ACL.
  27. Structured Context and High-Coverage Grammar for Conversational Question Answering over Knowledge Graphs. In EMNLP.
  28. Answering Conversational Questions on Structured Data without Logical Forms. In EMNLP-IJCNLP.
  29. UniK-QA: Unified Representations of Structured and Unstructured Knowledge for Open-Domain Question Answering. arXiv (2021).
  30. Context Transformer with Stacked Pointer Networks for Conversational Question Answering over Knowledge Graphs. In ESWC.
  31. UNIQORN: Unified Question Answering over RDF Knowledge Graphs and Natural Language Text. arXiv (2021).
  32. Reinforced History Backtracking for Conversational Question Answering. In AAAI, Vol. 35.
  33. Open-retrieval conversational question answering. In SIGIR.
  34. BERT with history answer embedding for conversational question answering. In SIGIR.
  35. Attentive history selection for conversational question answering. In CIKM.
  36. Exploring the limits of transfer learning with a unified text-to-text transformer. JMLR (2020).
  37. Question rewriting? Assessing its importance for conversational question answering. In ECIR.
  38. CoQA: A conversational question answering challenge. TACL (2019).
  39. Stephen Robertson and Hugo Zaragoza. 2009. The Probabilistic Relevance Framework: BM25 and Beyond. Foundations and Trends in Information Retrieval (2009).
  40. Recipes for Building an Open-Domain Chatbot. In EACL.
  41. Rishiraj Saha Roy and Avishek Anand. 2021. Question Answering for the Curated Web: Tasks and Methods in QA over Knowledge Bases and Text Collections. Morgan & Claypool Publishers.
  42. Complex sequential question answering: Towards learning to converse over linked question answer pairs with a knowledge graph. In AAAI.
  43. Denis Savenkov and Eugene Agichtein. 2016. When a knowledge base is not enough: Question answering over knowledge bases with external text data. In SIGIR.
  44. Entity Resolution in Open-domain Conversations. In NAACL.
  45. Multi-Task Learning for Conversational Question Answering over a Large-Scale Knowledge Base. In EMNLP-IJCNLP.
  46. What’s in a Name? Answer Equivalence For Open-Domain Question Answering. In EMNLP.
  47. YAGO: A core of semantic knowledge. In WWW.
  48. PullNet: Open Domain Question Answering with Iterative Retrieval on Knowledge Bases and Text. In EMNLP-IJCNLP.
  49. Open Domain Question Answering Using Early Fusion of Knowledge Bases and Text. In EMNLP.
  50. MultiModalQA: Complex Question Answering over Text, Tables and Images. In ICLR.
  51. Neural databases. arXiv (2020).
  52. Question rewriting for conversational question answering. In WSDM.
  53. Query resolution for conversational search with limited supervision. In SIGIR.
  54. Denny Vrandečić and Markus Krötzsch. 2014. Wikidata: A free collaborative knowledge base. CACM (2014).
  55. Improving Question Answering over Incomplete KBs with Knowledge-Aware Reader. In ACL.
  56. Question Answering on Freebase via Relation Extraction and Textual Evidence. In ACL.
  57. Learning to Generate Reformulation Actions for Scalable Conversational Query Understanding. In CIKM.
  58. TAT-QA: A Question Answering Benchmark on a Hybrid of Tabular and Textual Content in Finance. In ACL.
  59. Efficiency-aware Answering of Compositional Questions using Answer Type Prediction. In IJCNLP.
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (3)
  1. Philipp Christmann (9 papers)
  2. Rishiraj Saha Roy (23 papers)
  3. Gerhard Weikum (75 papers)
Citations (37)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now