Inference on Syntactic and Semantic Structures for Machine Comprehension

Authors: Chenrui Li, Yuanbin Wu, Man Lan

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

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments on the MCTest dataset demonstrate that the proposed models are highly competitive with state-of-the-art machine comprehension systems.
Researcher Affiliation Academia Chenrui Li,1 Yuanbin Wu,1, 2 Man Lan1, 2 1School of Computer Science and Software Engineering, East China Normal University 2Shanghai Key Laboratory of Multidimensional Information Processing {lcrr2011}@163.com, {ybwu, mlan}@cs.ecnu.edu.cn
Pseudocode Yes Algorithm 1 Message passing for the tree model ... Algorithm 2 Hidden Variable Perceptron
Open Source Code No The paper does not provide any explicit statement or link regarding the availability of its source code.
Open Datasets Yes The MCTest dataset (Richardson, Burges, and Renshaw 2013) contains 660 stories written for elementary grade school level students.
Dataset Splits Yes We use the official training, development and testing splitting.
Hardware Specification Yes The testing time is about 0.5s per question on a single CPU with 8G RAM.
Software Dependencies No The paper mentions 'Stanford Core NLP for coreference resolution', 'Stanford parser for dependency parsing', 'SEMAFOR frame-semantic parser', and 'word2vec' but does not specify version numbers for any of these software dependencies.
Experiment Setup Yes We tune the algorithm parameter on the development set, and set C = 0.001, M = 10.