Generating Diversified Comments via Reader-Aware Topic Modeling and Saliency Detection
Authors: Wei Wang, Piji Li, Hai-Tao Zheng13988-13996
AAAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experimental results on three datasets show that our framework outperforms existing baseline methods in terms of both automatic metrics and human evaluation. |
| Researcher Affiliation | Collaboration | Wei Wang1,2,* , Piji Li3, Hai-Tao Zheng1,2, 1Shenzhen International Graduate School, Tsinghua University 2Department of Computer Science and Technology, Tsinghua University 3Tencent AI Lab |
| Pseudocode | No | The paper describes methods and equations but does not present any clearly labeled pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide an explicit statement about releasing the source code for the described methodology, nor does it include a link to a code repository. |
| Open Datasets | Yes | Tencent Corpus is a Chinese dataset published in (Qin et al. 2018). Yahoo! News Corpus is an English dataset published in (Yang et al. 2019). Net Ease News Corpus is also a Chinese dataset crawled from Net Ease News4 and used in (Zheng et al. 2017). |
| Dataset Splits | Yes | Table 1: Statistics of the three datasets. Train Dev Test Tencent # News 191,502 5,000 1,610 Avg. # Cmts per News 5 27 27 Yahoo # News 152,355 5,000 3,160 Avg. # Cmts per News 7.7 20.5 20.5 Net Ease # News 75,287 5,000 2,500 Avg. # Cmts per News 22.7 22.5 22.5 |
| Hardware Specification | No | The paper does not specify any hardware details such as GPU models, CPU models, or specific machine configurations used for experiments. |
| Software Dependencies | No | The paper mentions 'Jie Ba', 'Stanford Core NLP', and 'Adam (Kingma and Ba 2014)' but does not provide specific version numbers for these or other software libraries/dependencies. |
| Experiment Setup | Yes | The embedding size is set to 256. For RNN based encoder, we use a two-layer Bi LSTM with hidden size 128. We use a two-layer LSTM with hidden size 256 as decoder. For self multi-head attention encoder, we use 4 heads and two layers. For CVAE and our topic modeling component, we set the size of latent variable to 64. For our method, λ1 and λ3 are set to 1, λ2 and λ4 are set to 0.5×10−3 and 0.2 respectively. We choose topic number K from set [10, 100, 1000], and we set K = 100 for Tencent dataset and K = 1000 for other two datasets. The dropout layer is inserted after LSTM layers of decoder and the dropout rate is set to 0.1 for regularization. The batch size is set to 128. We train the model using Adam (Kingma and Ba 2014) with learning rate 0.0005. We also clamp gradient values into the range [−8.0, 8.0] to avoid the exploding gradient problem (Pascanu, Mikolov, and Bengio 2013). |