Event Detection via Gated Multilingual Attention Mechanism

Authors: Jian Liu, Yubo Chen, Kang Liu, Jun Zhao

AAAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We have conducted extensive experiments on the ACE 2005 benchmark. Experimental results show that our approach significantly outperforms state-of-the-art methods.
Researcher Affiliation Academia Jian Liu,1,2 Yubo Chen,1 Kang Liu,1,2 Jun Zhao,1,2 1 National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, Beijing, 100190, China 2 University of Chinese Academy of Sciences, Beijing, 100049, China
Pseudocode No The paper describes the GMLATT framework and its components using text and equations, but it does not include a formal pseudocode block or algorithm.
Open Source Code No The paper does not provide any statement or link indicating the release of open-source code for the described methodology.
Open Datasets Yes We conduct experiments on the widely used ACE 2005 dataset. This corpus contains 599 documents annotated with 8 types and 33 subtypes of events. ... We use online machine translation service3 to obtain the parallel text in TARGET language. ... We concatenate the translated multilingual data with a 200k parallel English-Chinese corpus5 released by (Eisele and Chen 2010) to learn the alignments together. 3http://fanyi.baidu.com/, in our experiment 5http://opus.lingfil.uu.se/Multi UN.php
Dataset Splits Yes We use the same data separation as the previous works: 40 particular articles are used as the blind test set; 30 articles are used as the development set; and the remaining 529 articles are used for training.
Hardware Specification No The paper does not provide any specific details about the hardware used for running the experiments (e.g., GPU models, CPU types, memory).
Software Dependencies No The paper mentions software tools like GIZA++ and machine learning components like GRU and Bi-GRU, but it does not specify any version numbers for these software dependencies or libraries.
Experiment Setup No The paper mentions using 'mini-batch stochastic gradient descent (SGD)' and a 'dropout layer' and 'Negative sampling' for training and optimization, but it does not provide specific hyperparameter values such as learning rate, batch size, or dropout rate.