Skip to search formSkip to main contentSkip to account menu

Relation Classification via Recurrent Neural Network

    Dongxu ZhangDong Wang
    Computer Science
    arXiv.org
  • 5 August 2015
Experiments strongly indicates that the RNN-based model can deliver better performance on relation classification, and it is particularly capable of learning long-distance relation patterns, which makes it suitable for real-world applications where complicated expressions are often involved.
View PDF on arXiv (opens in a new tab)

Smoothing the Geometry of Probabilistic Box Embeddings

View Paper (opens in a new tab)

Word Embedding Perturbation for Sentence Classification

This technique report attempts several types of noise to perturb the input word embedding, such as Gaussian noise, Bernoulli noise, and adversarial noise, etc, to mitigate the overfitting problem of natural language.
View PDF on arXiv (opens in a new tab)

OpenKI: Integrating Open Information Extraction and Knowledge Bases with Relation Inference

This paper proposes OpenKI to handle sparsity of OpenIE extractions by performing instance-level inference: for each entity, it is proposed to encode the rich information in its neighborhood in both KB and OpenIE Extractions, and leverage this information in relation inference by exploring different methods of aggregation and attention.
View on ACL (opens in a new tab)

Search-Guided, Lightly-supervised Training of Structured Prediction Energy Networks

This paper uses efficient truncated randomized search in this reward function to train structured prediction energy networks (SPENs), which provide efficient test-time inference using gradient-based search on a smooth, learned representation of the score landscape, and have previously yielded state-of-the-art results in structured prediction.
View PDF on arXiv (opens in a new tab)

Learning from LDA Using Deep Neural Networks

A novel method is presented that uses LDA to supervise the training of a deep neural network (DNN), so that the DNN can approximate the costly LDA inference with less computation.
View PDF on arXiv (opens in a new tab)

Bitext Name Tagging for Cross-lingual Entity Annotation Projection

This paper focuses on the named entity recognition (NER) task and proposes a weakly-supervised framework to project entity annotations from English to IL through bitexts, which combines advantages of rule-based methods and deep learning methods.
View on ACL (opens in a new tab)

PKU @ CLSciSumm-17: Citation Contextualization

This report gives a brief introduction of the participants' participation in CL-SciSumm 2017 Task 1A, and demonstrates some data analysis and point out the difficulty of this task.
View Paper (opens in a new tab)

Capacity and Bias of Learned Geometric Embeddings for Directed Graphs

A novel variant of box embeddings is introduced that uses a learned smoothing parameter to achieve better representational capacity than vector models in low dimensions, while also avoiding performance saturation common to other geometric models in high dimensions.
View Paper (opens in a new tab)

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

ChemDisGene, a new dataset for training and evaluating multi-class multi-label biomedical relation extraction models, is introduced, which is both substantially larger and cleaner; it also includes annotations linking mentions to their entities.
View on ACL (opens in a new tab)
By clicking accept or continuing to use the site, you agree to the terms outlined in our Privacy Policy (opens in a new tab), Terms of Service (opens in a new tab), and Dataset License (opens in a new tab)