Entity-Relation Extraction as Multi-Turn Question Answering

By Xiaoya Li, Fan Yin, Zijun Sun, Xiayu Li, Arianna Yuan, Duo Chai, Mingxin Zhou, Jiwei Li. Annual Meeting of the Association for Computational Linguistics, 2019

388 citations recorded by the source corpus.

In this paper, we propose a new paradigm for the task of entity-relation extraction. We cast the task as a multi-turn question answering problem, i.e., the extraction of entities and elations is transformed to the task of identifying answer spans from the context. This multi-turn QA formalization comes with several key advantages: firstly, the question query encodes important information for the entity/relation class we want to identify; secondly, QA provides a natural way of jointly modeling entity and relation; and thirdly, it allows us to exploit the well developed machine reading comprehension (MRC) models. Experiments on the ACE and the CoNLL04 corpora demonstrate that the proposed paradigm significantly outperforms previous best models. We are able to obtain the state-of-the-art results on all of the ACE04, ACE05 and CoNLL04 datasets, increasing the SOTA results on the three datasets to 49.6 (+1.2), 60.3 (+0.7) and 69.2 (+1.4), respectively. Additionally, we construct and will release a newly developed dataset RESUME, which requires multi-step reasoning to construct entity dependencies, as opposed to the single-step dependency extraction in the triplet exaction in previous datasets. The proposed multi-turn QA model also achieves the best performance on the RESUME dataset.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/2c5ec74fb56fbfbceaa4cd5c8312ada4e2e19503

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

Question Answering is a Format; When is it Useful? (2019, methodology): “Das et al. (2019) and Li et al. (2019) achieve signiﬁcant improvement by converting the initial format of their data (entity tracking and relation extraction, respectively) to a QA format via question templates and using a QA model with no pretraining.”
A Unified MRC Framework for Named Entity Recognition (2019, methodology): “Additionally, Li et al. (2019) utilized a template-based procedure for constructing queries to extract semantic relations between entities and their queries lack diversity.”
Large language models are few-shot clinical information extractors (2022, methodology): “, first asking if a certain entity type exists in the input, before asking for a list (Li et al., 2019b; Wu et al., 2022) or (ii) using an output structure like the sequence tagging approach.”
TPLinker: Single-stage Joint Extraction of Entities and Relations Through Token Pair Linking (2020, background): “Feature-based works (Yu and Lam, 2010; Miwa and Sasaki, 2014) need a complicated process of feature engineering and heavily depend on NLP tools for feature extraction.”
Span-based Joint Entity and Relation Extraction with Transformer Pre-training (2019, methodology): “Unlike previous work based on BIO/BILOU labels (Bekoulis et al. 2018; Li et al. 2019; Nguyen and Verspoor 2019), a span-based approach can identify overlapping entities such as “codeine” within “codeine intoxication”.”
A Survey on Knowledge Graphs: Representation, Acquisition, and Applications (2020, background): “Some convert joint extraction into different problems such as sequence labeling via a novel tagging scheme [168] and multi-turn question answering [169].”
A Frustratingly Easy Approach for Entity and Relation Extraction (2021, methodology): “…models are publicly available at https: //github.com/princeton-nlp/PURE . that model these two tasks jointly (Li and Ji, 2014; Miwa and Bansal, 2016; Katiyar and Cardie, 2017; Zhang et al., 2017a; Li et al., 2019; Luan et al., 2018, 2019; Wadden et al., 2019; Lin et al., 2020; Wang and Lu, 2020).”
A Comprehensive Survey on Automatic Knowledge Graph Construction (2023, neutral): “[225] turned entity-relation tagging into a multi-turn question answering problem, leveraging the machine reading comprehensive (MRC) model for long-range semantics between entities.”
Structured Prediction as Translation between Augmented Natural Languages (2021, methodology): “Multi-turn QA (Li et al., 2019c) casts the problem as a multi-turn question answering task.”
Event Extraction as Machine Reading Comprehension (2020, methodology): “For example, Li et al. (2019a) use “Find organizations in the text” as a query command to find ORGANIZATION entity.”
Event Extraction by Answering (Almost) Natural Questions (2020, background): “Li et al. (2019b) casts entity-relation extraction as a multi-turn question answering task.”
Recent advances in deep learning based dialogue systems: a systematic survey (2021, methodology): “Lin et al. (2020a) combined a recurrent knowledge interactive decoder with a knowledge-aware pointer network to achieve both knowledgegrounded generation and knowledge copy.”