A Unified MRC Framework for Named Entity Recognition

By Xiaoya Li, Jingrong Feng, Yuxian Meng, Qinghong Han, Fei Wu, Jiwei Li. 2019

701 citations recorded by the source corpus.

The task of named entity recognition (NER) is normally divided into nested NER and flat NER depending on whether named entities are nested or not.Models are usually separately developed for the two tasks, since sequence labeling models, the most widely used backbone for flat NER, are only able to assign a single label to a particular token, which is unsuitable for nested NER where a token may be assigned several labels. In this paper, we propose a unified framework that is capable of handling both flat and nested NER tasks. Instead of treating the task of NER as a sequence labeling problem, we propose to formulate it as a machine reading comprehension (MRC) task. For example, extracting entities with the per label is formalized as extracting answer spans to the question “which person is mentioned in the text".This formulation naturally tackles the entity overlapping issue in nested NER: the extraction of two overlapping entities with different categories requires answering two independent questions. Additionally, since the query encodes informative prior knowledge, this strategy facilitates the process of entity extraction, leading to better performances for not only nested NER, but flat NER. We conduct experiments on both nested and flat NER datasets.Experiment results demonstrate the effectiveness of the proposed formulation. We are able to achieve a vast amount of performance boost over current SOTA models on nested NER datasets, i.e., +1.28, +2.55, +5.44, +6.37,respectively on ACE04, ACE05, GENIA and KBP17, along with SOTA results on flat NER datasets, i.e., +0.24, +1.95, +0.21, +1.49 respectively on English CoNLL 2003, English OntoNotes 5.0, Chinese MSRA and Chinese OntoNotes 4.0.

Subfields: Artificial Intelligence, Information Systems, Natural Language Processing

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

In Information Systems

In Natural Language Processing

Citations in the corpus

Finetuned Language Models Are Zero-Shot Learners (2021, methodology): “Other work has used QA-based task formulation for more-targeted applications including semantic role labeling (He et al., 2015), relation extraction (Levy et al., 2017), coreference resolution (Wu et al., 2020) and named entity recognition (Li et al., 2020) as question answering.”
Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processing (2021, background): “Later, this idea has been further explored in information extraction [70, 142] and text classification [12].”
Learning to Generate Task-Specific Adapters from Task Description (2021, background): “Several works study cross-task transfer by unifying the inputoutput format, e.g., relation extraction as machine reading comprehension (Levy et al., 2017), named entity recognition as machine reading comprehension (Li et al., 2020).”
Event Extraction by Answering (Almost) Natural Questions (2020, background): “Li et al. (2019b) casts entity-relation extraction as a multi-turn question answering task.”
GPT-NER: Named Entity Recognition via Large Language Models (2023, methodology): “• MRC-NER (Li et al., 2019a) formulates the NER task as a machine reading comprehen-”
SpanNER: Named Entity Re-/Recognition as Span Prediction (2021, methodology): “Although this is not the first work that formulates NER as a span prediction problem (Jiang et al., 2020; Ouchi et al., 2020; Yu et al., 2020; Li et al., 2020; Mengge et al., 2020), we contribute by (1) exploring how different design choices influence the performance of SPANNER and (2) interpreting ”
Zero-shot Learning by Generating Task-specific Adapters (2021, background): “, 2017), named entity recognition as machine reading comprehension (Li et al., 2020), coreference resolution as textual entailment (Yin et al.”
Interpretable Multi-dataset Evaluation for Named Entity Recognition (2020, background): “Orthogonally, some other works (Qian et al., 2018; Hu et al., 2020; Luo et al., 2020; Li et al., 2020; Lin et al., 2020) evaluate holistic metrics such as F1 across multiple datasets that differ in domain, language, or other characteristics (Sang and De Meulder, 2003; Collobert et al., 2011;…”
Structured Prediction as Translation between Augmented Natural Languages (2021, methodology): “In this line of work, BERT MRC (Li et al., 2019a) performs NER using two classiﬁcation modules to predict respectively the ﬁrst and the last tokens corresponding to an entity for a given input sentence.”
Few-NERD: A Few-shot Named Entity Recognition Dataset (2021, methodology): “, 2019a)) have become a strong encoder for NER (Li et al., 2020b).”
Example-Based Named Entity Recognition (2020, background): “For instance, (Wu et al., 2019) and (Li et al., 2019) have done this for coreference resolution and NER, respectively.”
Large language models are few-shot clinical information extractors (2022, positive): “To extract multiple spans simultaneously, Li et al. (2019b) and Li et al. (2019a)”
Query and Extract: Refining Event Extraction as Type-oriented Binary Decoding (2021, background): “Moreover, compared with the QA-based083 methods (Du and Cardie, 2020; Liu et al., 2020;084 Li et al., 2020) that can also conduct zero-shot ar-085 gument extraction, our approach does not require086 creating high-quality questions for argument roles087 or multi-time encoding for different argument…”
Description Based Text Classification with Reinforcement Learning (2020, methodology): “Inspired by the current trend of formalizing NLP problems as question answering tasks (Levy et al., 2017; McCann et al., 2018; Li et al., 2019a,b; Gardner et al., 2019; Raffel et al., 2019), we pro-pose a new framework for text classiﬁcation by formalizing it as a SQuAD-style machine reading…”
A Survey on Knowledge Graphs: Representation, Acquisition, and Applications (2020, background): “[106] formulated flat and nested NER as a unified machine reading comprehension framework”
Dice Loss for Data-imbalanced NLP Tasks (2019, methodology): “• Glyce-BERT: • BERT-MRC: The current SOTA model for both Chinese and English NER datasets proposed by Li et al. (2019), which formulate NER as machine reading comprehension task.”
DeepStruct: Pretraining of Language Models for Structure Prediction (2022, methodology): “We use the preprocessed version of this dataset from (Li et al., 2020a).”