EMNLP Findings2023
From Speculation Detection to Trustworthy Relational Tuples in Information Extraction
Kuicai Dong, Aixin Sun, Jung-Jae Kim, Xiaoli Li
Abstract
Speculation detection is an important NLP task to understand text factuality. However, the extracted speculative information (e.g., speculative polarity, cue, and scope) lacks structure and poses challenges for direct utilization in downstream tasks. Open Information Extraction (OIE), on the other hand, extracts structured tuples as facts, without examining the certainty of these tuples. Bridging this gap between speculation detection and information extraction becomes imperative to generate structured speculative information and trustworthy relational tuples. Existing studies on speculation detection are defined at sentence level; but even if a sentence is determined to be speculative, not all factual tuples extracted from it are speculative. In this paper, we propose to study speculations in OIE tuples and determine whether a tuple is speculative. We formally define the research problem of tuple-level speculation detection. We then conduct detailed analysis on the LSOIE dataset which provides labels for speculative tuples. Lastly, we propose a baseline model SpecTup for this new research task. 1
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
- DocOIE: A Document-level Context-Aware Dataset for OpenIE cites OpenIE6: Iterative Grid Labeling and Coordination Analysis for Open Information Extraction
- Syntactic Multi-view Learning for Open Information Extraction cites DocOIE: A Document-level Context-Aware Dataset for OpenIE
- Syntactic Multi-view Learning for Open Information Extraction cites OpenIE6: Iterative Grid Labeling and Coordination Analysis for Open Information Extraction
- Syntactic Multi-view Learning for Open Information Extraction cites DetIE: Multilingual Open Information Extraction Inspired by Object Detection
- DetIE: Multilingual Open Information Extraction Inspired by Object Detection cites OpenIE6: Iterative Grid Labeling and Coordination Analysis for Open Information Extraction
Builds on6
- Composition-based Multi-Relational Graph Convolutional NetworksShikhar Vashishth, Soumya Sanyal, Vikram Nitin, Partha P. TalukdarICLR 2020 · 1,105 citations
- DetIE: Multilingual Open Information Extraction Inspired by Object DetectionMichael Vasilkovsky, Anton Alekseev, Valentin Malykh, Ilya Shenbin et al.AAAI 2022 · 24 citations
- OpenIE6: Iterative Grid Labeling and Coordination Analysis for Open Information ExtractionKeshav Kolluru, Vaibhav Adlakha, Samarth Aggarwal, Mausam et al.EMNLP 2020 · 13 citations
- DocOIE: A Document-level Context-Aware Dataset for OpenIEKuicai Dong, Yilin Zhao, Aixin Sun, Jung-Jae Kim et al.ACL Findings 2021 · 12 citations
- Syntactic Multi-view Learning for Open Information ExtractionKuicai Dong, Aixin Sun, Jung-Jae Kim, Xiaoli LiEMNLP 2022 · 6 citations
Related papers
- Open Information Extraction via ChunksKuicai Dong, Aixin Sun, Jung-Jae Kim, Xiaoli LiEMNLP 2023 · 4 citations
- Abstractive Open Information ExtractionKevin Pei, Ishan Jindal, Kevin Chen-Chuan ChangEMNLP 2023
- Semi-Open Information ExtractionBowen Yu, Zhenyu Zhang, Jiawei Sheng, Tingwen Liu et al.WWW 2021 · 29 citations
- Syntactically Rich Discriminative Training: An Effective Method for Open Information ExtractionFrank Mtumbuka, Thomas LukasiewiczEMNLP 2022 · 1 citation
- Maximal Clique Based Non-Autoregressive Open Information ExtractionBowen Yu, Yucheng Wang, Tingwen Liu, Hongsong Zhu et al.EMNLP 2021 · 14 citations
