Utilizing Non-Parallel Text for Style Transfer by Making Partial Comparisons

Authors: Di Yin, Shujian Huang, Xin-Yu Dai, Jiajun Chen

IJCAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We perform experiments to compare our method to other existing approaches on two review datasets. Both automatic and manual evaluations show that our approach can significantly improve the performance of existing adversarial methods, and outperforms most state-of-the-art models.
Researcher Affiliation Academia Di Yin, Shujian Huang , Xin-Yu Dai and Jiajun Chen National Key Laboratory for Novel Software Technology Nanjing University, Nanjing 210023, China yind@smail.nju.edu.cn, {huangsj, daixinyu, chenjj}@nju.edu.cn
Pseudocode Yes Algorithm 1 Making partial comparisons
Open Source Code Yes Our code and data will be available on Github1. ... 1https://github.com/yd1996/Partial Comparison
Open Datasets Yes Datasets We conduct experiments on the Yelp review dataset and Amazon review dataset (Table 1) released by [Li et al., 2018a]
Dataset Splits Yes Table 1: Statistics of the datasets. ... Dataset Attributes Train Dev Test ... Hyper-parameters are selected based on the validation set
Hardware Specification Yes We implement our model based on Py Torch4 and use four NVIDIA GTX1080Ti graphic cards for learning.
Software Dependencies No The paper mentions 'Py Torch' and 'Word2Vec' but does not specify their exact version numbers. It also refers to optimization algorithms like Adam and RMSprop, which are not software dependencies with specific versions.
Experiment Setup Yes The encoder E and the generator G are single-layer LSTMRNNs with input dimension of 300 and hidden dimensions of 350. The dimension of style embedding is 50. We use a batch size of 160, which contains 80 samples from X and Y respectively. Hyper-parameters are selected based on the validation set, and we use grid search to pick the best parameters. The learning rate is selected from [1e 4, 2e 4, 5e 4, 1e 3], and the weights of each term in the training objective (λz, λc, λs, λlm) are all selected from [0.1, 0.5, 1.0, 2.0, 5.0].