PyTorchKR Papers

최신 논문

총 1,251편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2204.13516공식 코드 · Papers with Code

What do You Mean by Relation Extraction? A Survey on Datasets and Study on Scientific Relation Classification

Over the last five years, research on Relation Extraction (RE) witnessed extensive progress with many new dataset releases. At the same time, setup clarity has decreased, contributing to increased difficulty of reliable empirical evaluation (Taill\'e et al., 2020). In this paper, we provide a comprehensive survey of RE datasets, and revisit the task definition and its adoption by the community. We find that cross-dataset and cross-domain setups are particularly lacking. We present an empirical study on scientific Relation Classification across two datasets. Despite large data overlap, our analysis reveals substantial discrepancies in annotation. Annotation discrepancies strongly impact Relation Classification performance, explaining large drops in cross-dataset evaluations. Variation within further sub-domains exists but impacts Relation Classification only to limited degrees. Overall, our study calls for more rigour in reporting setups in RE and evaluation across multiple test sets.

arXiv:2204.12710

CREER: A Large-Scale Corpus for Relation Extraction and Entity Recognition

We describe the design and use of the CREER dataset, a large corpus annotated with rich English grammar and semantic attributes. The CREER dataset uses the Stanford CoreNLP Annotator to capture rich language structures from Wikipedia plain text. This dataset follows widely used linguistic and semantic annotations so that it can be used for not only most natural language processing tasks but also scaling the dataset. This large supervised dataset can serve as the basis for improving the performance of NLP tasks in the future. We publicize the dataset through the link: https://140.116.82.111/share.cgi?ssid=000dOJ4

arXiv:2204.12679공식 코드 · Papers with Code

Document-Level Relation Extraction with Sentences Importance Estimation and Focusing

Document-level relation extraction (DocRE) aims to determine the relation between two entities from a document of multiple sentences. Recent studies typically represent the entire document by sequence- or graph-based models to predict the relations of all entity pairs. However, we find that such a model is not robust and exhibits bizarre behaviors: it predicts correctly when an entire test document is fed as input, but errs when non-evidence sentences are removed. To this end, we propose a Sentence Importance Estimation and Focusing (SIEF) framework for DocRE, where we design a sentence importance score and a sentence focusing loss, encouraging DocRE models to focus on evidence sentences. Experimental results on two domains show that our SIEF not only improves overall performance, but also makes DocRE models more robust. Moreover, SIEF is a general framework, shown to be effective when combined with a variety of base DocRE models.

arXiv:2204.11424공식 코드 · Papers with Code

It Takes Two Flints to Make a Fire: Multitask Learning of Neural Relation and Explanation Classifiers

We propose an explainable approach for relation extraction that mitigates the tension between generalization and explainability by jointly training for the two goals. Our approach uses a multi-task learning architecture, which jointly trains a classifier for relation extraction, and a sequence model that labels words in the context of the relation that explain the decisions of the relation classifier. We also convert the model outputs to rules to bring global explanations to this approach. This sequence model is trained using a hybrid strategy: supervised, when supervision from pre-existing patterns is available, and semi-supervised otherwise. In the latter situation, we treat the sequence model's labels as latent variables, and learn the best assignment that maximizes the performance of the relation classifier. We evaluate the proposed approach on the two datasets and show that the sequence model provides labels that serve as accurate explanations for the relation classifier's decisions, and, importantly, that the joint training generally improves the performance of the relation classifier. We also evaluate the performance of the generated rules and show that the new rules are great add-on to the manual rules and bring the rule-based system much closer to the neural models.

arXiv:2204.10360

Decorate the Examples: A Simple Method of Prompt Design for Biomedical Relation Extraction

Relation extraction is a core problem for natural language processing in the biomedical domain. Recent research on relation extraction showed that prompt-based learning improves the performance on both fine-tuning on full training set and few-shot training. However, less effort has been made on domain-specific tasks where good prompt design can be even harder. In this paper, we investigate prompting for biomedical relation extraction, with experiments on the ChemProt dataset. We present a simple yet effective method to systematically generate comprehensive prompts that reformulate the relation extraction task as a cloze-test task under a simple prompt formulation. In particular, we experiment with different ranking scores for prompt selection. With BioMed-RoBERTa-base, our results show that prompting-based fine-tuning obtains gains by 14.21 F1 over its regular fine-tuning baseline, and 1.14 F1 over SciFive-Large, the current state-of-the-art on ChemProt. Besides, we find prompt-based learning requires fewer training examples to make reasonable predictions. The results demonstrate the potential of our methods in such a domain-specific relation extraction task.

arXiv:2204.09952

Recovering Patient Journeys: A Corpus of Biomedical Entities and Relations on Twitter (BEAR)

Text mining and information extraction for the medical domain has focused on scientific text generated by researchers. However, their direct access to individual patient experiences or patient-doctor interactions can be limited. Information provided on social media, e.g., by patients and their relatives, complements the knowledge in scientific text. It reflects the patient's journey and their subjective perspective on the process of developing symptoms, being diagnosed and offered a treatment, being cured or learning to live with a medical condition. The value of this type of data is therefore twofold: Firstly, it offers direct access to people's perspectives. Secondly, it might cover information that is not available elsewhere, including self-treatment or self-diagnoses. Named entity recognition and relation extraction are methods to structure information that is available in unstructured text. However, existing medical social media corpora focused on a comparably small set of entities and relations and particular domains, rather than putting the patient into the center of analyses. With this paper we contribute a corpus with a rich set of annotation layers following the motivation to uncover and model patients' journeys and experiences in more detail. We label 14 entity classes (incl. environmental factors, diagnostics, biochemical processes, patients' quality-of-life descriptions, pathogens, medical conditions, and treatments) and 20 relation classes (e.g., prevents, influences, interactions, causes) most of which have not been considered before for social media data. The publicly available dataset consists of 2,100 tweets with approx. 6,000 entity and 3,000 relation annotations. In a corpus analysis we find that over 80 % of documents contain relevant entities. Over 50 % of tweets express relations which we consider essential for uncovering patients' narratives about their journeys.

arXiv:2204.09851

A Masked Image Reconstruction Network for Document-level Relation Extraction

Document-level relation extraction aims to extract relations among entities within a document. Compared with its sentence-level counterpart, Document-level relation extraction requires inference over multiple sentences to extract complex relational triples. Previous research normally complete reasoning through information propagation on the mention-level or entity-level document-graphs, regardless of the correlations between the relationships. In this paper, we propose a novel Document-level Relation Extraction model based on a Masked Image Reconstruction network (DRE-MIR), which models inference as a masked image reconstruction problem to capture the correlations between relationships. Specifically, we first leverage an encoder module to get the features of entities and construct the entity-pair matrix based on the features. After that, we look on the entity-pair matrix as an image and then randomly mask it and restore it through an inference module to capture the correlations between the relationships. We evaluate our model on three public document-level relation extraction datasets, i.e. DocRED, CDR, and GDA. Experimental results demonstrate that our model achieves state-of-the-art performance on these three datasets and has excellent robustness against the noises during the inference process.

arXiv:2204.07980공식 코드 · Papers with Code

Does Recommend-Revise Produce Reliable Annotations? An Analysis on Missing Instances in DocRED

DocRED is a widely used dataset for document-level relation extraction. In the large-scale annotation, a \textit{recommend-revise} scheme is adopted to reduce the workload. Within this scheme, annotators are provided with candidate relation instances from distant supervision, and they then manually supplement and remove relational facts based on the recommendations. However, when comparing DocRED with a subset relabeled from scratch, we find that this scheme results in a considerable amount of false negative samples and an obvious bias towards popular entities and relations. Furthermore, we observe that the models trained on DocRED have low recall on our relabeled dataset and inherit the same bias in the training data. Through the analysis of annotators' behaviors, we figure out the underlying reason for the problems above: the scheme actually discourages annotators from supplementing adequate instances in the revision phase. We appeal to future research to take into consideration the issues with the recommend-revise scheme when designing new models and annotation schemes. The relabeled dataset is released at \url{https://github.com/AndrewZhe/Revisit-DocRED}, to serve as a more reliable test set of document RE models.

arXiv:2204.08387공식 코드 · Papers with Code

LayoutLMv3: Pre-training for Document AI with Unified Text and Image Masking

Self-supervised pre-training techniques have achieved remarkable progress in Document AI. Most multimodal pre-trained models use a masked language modeling objective to learn bidirectional representations on the text modality, but they differ in pre-training objectives for the image modality. This discrepancy adds difficulty to multimodal representation learning. In this paper, we propose \textbf{LayoutLMv3} to pre-train multimodal Transformers for Document AI with unified text and image masking. Additionally, LayoutLMv3 is pre-trained with a word-patch alignment objective to learn cross-modal alignment by predicting whether the corresponding image patch of a text word is masked. The simple unified architecture and training objectives make LayoutLMv3 a general-purpose pre-trained model for both text-centric and image-centric Document AI tasks. Experimental results show that LayoutLMv3 achieves state-of-the-art performance not only in text-centric tasks, including form understanding, receipt understanding, and document visual question answering, but also in image-centric tasks such as document image classification and document layout analysis. The code and models are publicly available at \url{https://aka.ms/layoutlmv3}.

arXiv:2204.07150공식 코드 · Papers with Code

FREDA: Flexible Relation Extraction Data Annotation

To effectively train accurate Relation Extraction models, sufficient and properly labeled data is required. Adequately labeled data is difficult to obtain and annotating such data is a tricky undertaking. Previous works have shown that either accuracy has to be sacrificed or the task is extremely time-consuming, if done accurately. We are proposing an approach in order to produce high-quality datasets for the task of Relation Extraction quickly. Neural models, trained to do Relation Extraction on the created datasets, achieve very good results and generalize well to other datasets. In our study, we were able to annotate 10,022 sentences for 19 relations in a reasonable amount of time, and trained a commonly used baseline model for each relation.

arXiv:2204.06584공식 코드 · Papers with Code

A Distant Supervision Corpus for Extracting Biomedical Relationships Between Chemicals, Diseases and Genes

We introduce ChemDisGene, a new dataset for training and evaluating multi-class multi-label document-level biomedical relation extraction models. Our dataset contains 80k biomedical research abstracts labeled with mentions of chemicals, diseases, and genes, portions of which human experts labeled with 18 types of biomedical relationships between these entities (intended for evaluation), and the remainder of which (intended for training) has been distantly labeled via the CTD database with approximately 78\% accuracy. In comparison to similar preexisting datasets, ours is both substantially larger and cleaner; it also includes annotations linking mentions to their entities. We also provide three baseline deep neural network relation extraction models trained and evaluated on our new dataset.

arXiv:2204.04263

BioRED: A Rich Biomedical Relation Extraction Dataset

Automated relation extraction (RE) from biomedical literature is critical for many downstream text mining applications in both research and real-world settings. However, most existing benchmarking datasets for bio-medical RE only focus on relations of a single type (e.g., protein-protein interactions) at the sentence level, greatly limiting the development of RE systems in biomedicine. In this work, we first review commonly used named entity recognition (NER) and RE datasets. Then we present BioRED, a first-of-its-kind biomedical RE corpus with multiple entity types (e.g., gene/protein, disease, chemical) and relation pairs (e.g., gene-disease; chemical-chemical) at the document level, on a set of 600 PubMed abstracts. Further, we label each relation as describing either a novel finding or previously known background knowledge, enabling automated algorithms to differentiate between novel and background information. We assess the utility of BioRED by benchmarking several existing state-of-the-art methods, including BERT-based models, on the NER and RE tasks. Our results show that while existing approaches can reach high performance on the NER task (F-score of 89.3%), there is much room for improvement for the RE task, especially when extracting novel relations (F-score of 47.7%). Our experiments also demonstrate that such a rich dataset can successfully facilitate the development of more accurate, efficient, and robust RE systems for biomedicine. The BioRED dataset and annotation guideline are freely available at https://ftp.ncbi.nlm.nih.gov/pub/lu/BioRED/.

arXiv:2204.04779공식 코드 · Papers with Code

MedDistant19: Towards an Accurate Benchmark for Broad-Coverage Biomedical Relation Extraction

Relation extraction in the biomedical domain is challenging due to the lack of labeled data and high annotation costs, needing domain experts. Distant supervision is commonly used to tackle the scarcity of annotated data by automatically pairing knowledge graph relationships with raw texts. Such a pipeline is prone to noise and has added challenges to scale for covering a large number of biomedical concepts. We investigated existing broad-coverage distantly supervised biomedical relation extraction benchmarks and found a significant overlap between training and test relationships ranging from 26% to 86%. Furthermore, we noticed several inconsistencies in the data construction process of these benchmarks, and where there is no train-test leakage, the focus is on interactions between narrower entity types. This work presents a more accurate benchmark MedDistant19 for broad-coverage distantly supervised biomedical relation extraction that addresses these shortcomings and is obtained by aligning the MEDLINE abstracts with the widely used SNOMED Clinical Terms knowledge base. Lacking thorough evaluation with domain-specific language models, we also conduct experiments validating general domain relation extraction findings to biomedical relation extraction.

arXiv:2204.04437공식 코드 · Papers with Code

Modeling Multi-Granularity Hierarchical Features for Relation Extraction

Relation extraction is a key task in Natural Language Processing (NLP), which aims to extract relations between entity pairs from given texts. Recently, relation extraction (RE) has achieved remarkable progress with the development of deep neural networks. Most existing research focuses on constructing explicit structured features using external knowledge such as knowledge graph and dependency tree. In this paper, we propose a novel method to extract multi-granularity features based solely on the original input sentences. We show that effective structured features can be attained even without external knowledge. Three kinds of features based on the input sentences are fully exploited, which are in entity mention level, segment level, and sentence level. All the three are jointly and hierarchically modeled. We evaluate our method on three public benchmarks: SemEval 2010 Task 8, Tacred, and Tacred Revisited. To verify the effectiveness, we apply our method to different encoders such as LSTM and BERT. Experimental results show that our method significantly outperforms existing state-of-the-art models that even use external knowledge. Extensive analyses demonstrate that the performance of our model is contributed by the capture of multi-granularity features and the model of their hierarchical structure. Code and data are available at \url{https://github.com/xnliang98/sms}.

arXiv:2204.01098공식 코드 · Papers with Code

A sequence-to-sequence approach for document-level relation extraction

Motivated by the fact that many relations cross the sentence boundary, there has been increasing interest in document-level relation extraction (DocRE). DocRE requires integrating information within and across sentences, capturing complex interactions between mentions of entities. Most existing methods are pipeline-based, requiring entities as input. However, jointly learning to extract entities and relations can improve performance and be more efficient due to shared parameters and training steps. In this paper, we develop a sequence-to-sequence approach, seq2rel, that can learn the subtasks of DocRE (entity extraction, coreference resolution and relation extraction) end-to-end, replacing a pipeline of task-specific components. Using a simple strategy we call entity hinting, we compare our approach to existing pipeline-based methods on several popular biomedical datasets, in some cases exceeding their performance. We also report the first end-to-end results on these datasets for future comparison. Finally, we demonstrate that, under our model, an end-to-end approach outperforms a pipeline-based approach. Our code, data and trained models are available at {\url{https://github.com/johngiorgi/seq2rel}}. An online demo is available at {\url{https://share.streamlit.io/johngiorgi/seq2rel/main/demo.py}}.

arXiv:2204.00255

NC-DRE: Leveraging Non-entity Clue Information for Document-level Relation Extraction

Document-level relation extraction (RE), which requires reasoning on multiple entities in different sentences to identify complex inter-sentence relations, is more challenging than sentence-level RE. To extract the complex inter-sentence relations, previous studies usually employ graph neural networks (GNN) to perform inference upon heterogeneous document-graphs. Despite their great successes, these graph-based methods, which normally only consider the words within the mentions in the process of building graphs and reasoning, tend to ignore the non-entity clue words that are not in the mentions but provide important clue information for relation reasoning. To alleviate this problem, we treat graph-based document-level RE models as an encoder-decoder framework, which typically uses a pre-trained language model as the encoder and a GNN model as the decoder, and propose a novel graph-based model NC-DRE that introduces decoder-to-encoder attention mechanism to leverage Non-entity Clue information for Document-level Relation Extraction.

arXiv:2203.15827공식 코드 · Papers with Code

LinkBERT: Pretraining Language Models with Document Links

Language model (LM) pretraining can learn various knowledge from text corpora, helping downstream tasks. However, existing methods such as BERT model a single document, and do not capture dependencies or knowledge that span across documents. In this work, we propose LinkBERT, an LM pretraining method that leverages links between documents, e.g., hyperlinks. Given a text corpus, we view it as a graph of documents and create LM inputs by placing linked documents in the same context. We then pretrain the LM with two joint self-supervised objectives: masked language modeling and our new proposal, document relation prediction. We show that LinkBERT outperforms BERT on various downstream tasks across two domains: the general domain (pretrained on Wikipedia with hyperlinks) and biomedical domain (pretrained on PubMed with citation links). LinkBERT is especially effective for multi-hop reasoning and few-shot QA (+5% absolute improvement on HotpotQA and TriviaQA), and our biomedical LinkBERT sets new states of the art on various BioNLP tasks (+7% on BioASQ and USMLE). We release our pretrained models, LinkBERT and BioLinkBERT, as well as code and data at https://github.com/michiyasunaga/LinkBERT.

arXiv:2203.15323

Improving Persian Relation Extraction Models by Data Augmentation

Relation extraction that is the task of predicting semantic relation type between entities in a sentence or document is an important task in natural language processing. Although there are many researches and datasets for English, Persian suffers from sufficient researches and comprehensive datasets. The only available Persian dataset for this task is PERLEX, which is a Persian expert-translated version of the SemEval-2010-Task-8 dataset. In this paper, we present our augmented dataset and the results and findings of our system, participated in the Persian relation Extraction shared task of NSURL 2021 workshop. We use PERLEX as the base dataset and enhance it by applying some text preprocessing steps and by increasing its size via data augmentation techniques to improve the generalization and robustness of applied models. We then employ two different models including ParsBERT and multilingual BERT for relation extraction on the augmented PERLEX dataset. Our best model obtained 64.67% of Macro-F1 on the test phase of the contest and it achieved 83.68% of Macro-F1 on the test set of PERLEX.

arXiv:2203.13953

A Densely Connected Criss-Cross Attention Network for Document-level Relation Extraction

Document-level relation extraction (RE) aims to identify relations between two entities in a given document. Compared with its sentence-level counterpart, document-level RE requires complex reasoning. Previous research normally completed reasoning through information propagation on the mention-level or entity-level document-graph, but rarely considered reasoning at the entity-pair-level.In this paper, we propose a novel model, called Densely Connected Criss-Cross Attention Network (Dense-CCNet), for document-level RE, which can complete logical reasoning at the entity-pair-level. Specifically, the Dense-CCNet performs entity-pair-level logical reasoning through the Criss-Cross Attention (CCA), which can collect contextual information in horizontal and vertical directions on the entity-pair matrix to enhance the corresponding entity-pair representation. In addition, we densely connect multiple layers of the CCA to simultaneously capture the features of single-hop and multi-hop logical reasoning.We evaluate our Dense-CCNet model on three public document-level RE datasets, DocRED, CDR, and GDA. Experimental results demonstrate that our model achieves state-of-the-art performance on these three datasets.

arXiv:2203.14710공식 코드 · Papers with Code

Hierarchical Transformer Model for Scientific Named Entity Recognition

The task of Named Entity Recognition (NER) is an important component of many natural language processing systems, such as relation extraction and knowledge graph construction. In this work, we present a simple and effective approach for Named Entity Recognition. The main idea of our approach is to encode the input subword sequence with a pre-trained transformer such as BERT, and then, instead of directly classifying the word labels, another layer of transformer is added to the subword representation to better encode the word-level interaction. We evaluate our approach on three benchmark datasets for scientific NER, particularly in the computer science and biomedical domains. Experimental results show that our model outperforms the current state-of-the-art on SciERC and TDM datasets without requiring external resources or specific data augmentation. Code is available at \url{https://github.com/urchade/HNER}.