Generating Thematic Chinese Poetry using Conditional Variational Autoencoders with Hybrid Decoders
Authors: Xiaopeng Yang, Xiaowen Lin, Shunda Suo, Ming Li
IJCAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Tests show that the generated poems by our approach are mostly satisfying with regulated rules and consistent themes, and 73.42% of them receive an Overall score no less than 3 (the highest score is 5)., We conduct both automatic and human evaluation to verify the feasibility and availability of our proposed Chinese poetry generation approach. |
| Researcher Affiliation | Academia | Xiaopeng Yang, Xiaowen Lin, Shunda Suo, and Ming Li David R. Cheriton School of Computer Science, Faculty of Mathematics, University of Waterloo Waterloo, ON, Canada N2L 3G1 {x335yang, x65lin, sdsuo, mli}@uwaterloo.ca |
| Pseudocode | No | The paper describes the approach and optimization but does not include any explicitly labeled 'Pseudocode' or 'Algorithm' blocks. |
| Open Source Code | No | The paper does not provide any statement about releasing source code for the described methodology or a link to a code repository. |
| Open Datasets | No | Two large-scale datasets are used in our experiments. The first dataset is a Chinese poem corpus (CPC) containing 284,899 traditional Chinese poems in various genres... We use this dataset to train the wordembedding for Chinese characters. Since we focus on generating quatrains... we filter 76,305 quatrains from CPC, named as Chinese quatrain corpus (CQC), to train the neural network model. The paper does not provide a specific link, DOI, or formal citation for public access to these datasets. |
| Dataset Splits | Yes | Specifically, we randomly choose 2,000 poems for validation, 1,000 poems for testing, and other non-overlap ones for training. |
| Hardware Specification | No | The paper does not specify any hardware details (e.g., GPU/CPU models, memory, or cloud resources) used for running the experiments. |
| Software Dependencies | No | The paper mentions optimization algorithms like Ada Delta, but does not provide specific software dependencies with version numbers (e.g., programming language versions, library versions, or specific framework versions). |
| Experiment Setup | Yes | The dimension of word-embedding vectors is set to 128. The recurrent hidden layers of the encoder and the RNN part of the hybrid decoder contain 128 hidden units, and the number of layers is both set to 4. We use 3 deconvolutional layers with the Re LU non-linearity in the de CNN part of the hybrid decoder. The kernel size is set to 3 and the stride is 2. The number of feature maps is [256, 128, 64] for each layer respectively. The weighting parameter α is set to 0.6. We use 64-dimensional latent variables. Parameters of our model were randomly initialized over a uniform distribution with support [-0.08,0.08]. The model is trained using the Ada Delta algorithm [Zeiler, 2012], where the mini-batch is set to 64 and the learning rate is 0.001. The dropout technique [Srivastava et al., 2014] is also adopted and the dropout rate is set to 0.2. |